This paper presents a novel approach for indoor acoustic source localization using microphone arrays and based on a Convolutional Neural Network (CNN). The proposed solution is, to the best of our knowledge, the first published work in which the CNN is designed to directly estimate the three dimensional position of an acoustic source, using the raw audio signal as the input information avoiding the use of hand crafted audio features. Given the limited amount of available localization data, we propose in this paper a training strategy based on two steps. We first train our network using semi-synthetic data, generated from close talk speech recordings, and where we simulate the time delays and distortion suffered in the signal that propagates from the source to the array of microphones. We then fine tune this network using a small amount of real data. Our experimental results show that this strategy is able to produce networks that significantly improve existing localization methods based on SRP-PHAT strategies. In addition, our experiments show that our CNN method exhibits better resistance against varying gender of the speaker and different window sizes compared with the other methods.

This article presents a study on brain tumor detection using the VGG-16 model, a convolutional neural network known for its performance in computer vision tasks. The aim of the study is to classify magnetic resonance imaging (MRI) images and accurately identify the presence of brain tumors. The dataset used consists of brain tumor MRI images, categorized into two classes: "NO" (no tumor) and "YES" (tumor). The methodology involves setting up the environment, importing and preprocessing the data, building the VGG-16 model, and evaluating its performance using metrics such as accuracy, precision, and recall. The results demonstrate an accuracy of approximately 88% on the validation set and 80% on the test set, indicating the potential of the VGG-16 model in supporting healthcare professionals in diagnosing brain tumors. The study contributes to the field of medical image analysis and offers insights into the application of deep learning for brain tumor detection.

Radio Frequency Interference (RFI) detection and characterization play a critical role to in ensuring the security of all wireless communication networks. Advances in Machine Learning (ML) have led to the deployment of many robust techniques dealing with various types of RFI. To sidestep an unavoidable complicated feature extraction step in ML, this paper proposes an efficient end-to-end method using the latest advances in deep learning to extract the appropriate features of the RFI signal. Moreover, this study utilizes the benefits of transfer learning to determine both the type of received RFI signals and their modulation types. To this end, the scalogram of the received signals is used as the input of the pre-trained convolutional neural networks (CNN), followed by a fully-connected classifier. This study considers a digital video stream as the signal of interest (SoI), transmitted in a real-time satellite-to-ground communication using DVB-S2 standards. To create the RFI dataset, the SoI is combined with three well-known jammers namely, continuous-wave interference (CWI), multi- continuous-wave interference (MCWI), and chirp interference (CI). This study investigated four well-known pre-trained CNN architectures, namely, AlexNet, VGG-16, GoogleNet, and ResNet-18, for the feature extraction to recognize the visual RFI patterns directly from pixel images with minimal preprocessing. Moreover, the robustness of the proposed classifiers is evaluated by the data generated at different signal to noise ratios (SNR).

The COVID-19 is a highly contagious viral infection which played havoc on everyone's life in many different ways. According to the world health organization and scientists, more testing potentially helps governments and disease control organizations in containing the spread of the virus. The use of chest radiographs is one of the early screening tests to determine the onset of disease, as the infection affects the lungs severely. This study will investigate and automate the process of testing by using state-of-the-art CNN classifiers to detect the COVID19 infection. However, the viral could of many different types; therefore, we only regard for COVID19 while the other viral infection types are treated as non-COVID19 in the radiographs of various viral infections. The classification task is challenging due to the limited number of scans available for COVID19 and the minute variations in the viral infections. We aim to employ current state-of-the-art CNN architectures, compare their results, and determine whether deep learning algorithms can handle the crisis appropriately. All trained models are available at https://github.com/saeed-anwar/COVID19-Baselines

A topic of growing interest in urban remote sensing is the automated extraction of geometrical building information for 3D city modeling. Roof geometry information is useful for applications such as urban planning, solar potential estimation and telecommunication installation planning, and wind flow simulations for pollutant diffusion analysis. Recent research has proven that the advance in remote sensing technologies and deep learning methods offer the prospects of deriving the roof structure information accurately and efficiently. In this study, we propose a Vectorized Roof Extractor- method based on Fully Convolutional Networks (FCNs) and advanced polygonization method to extract roof structure from aerial imagery and a normalized Digital Surface Models (nDSM) in a regularized vector format. The roof structure consists of building outlines, external edges of the building roof, inner rooflines, internal intersections of the main roof planes. The methodology is comprised of segmentation, vectorization and post-processing for outer rooflines, external edges of the building roof, and inner rooflines, and internal intersections of the main roof planes. For the comparison, we adapt the Frame field Learning (FFL) method originally designed to extract building polygons [1]. Our experiments are conducted on a custom data set derived for the city of Enschede, The Netherlands, using aerial imagery, nDSM and manually digitized training polygons. The results show that the proposed Vectorized Roof Extractor outperformed adapted FFL on PoLiS distance with values of 3.5 m and 1.2 m for outlines and inner rooflines, respectively. Furthermore, the model surpassed the adapted FFL on PoLiS-thresholded F-score for outlines and inner rooflines, with 0.31 and 0.57, respectively. The Vectorized Roof Extractor produced adequate visual results, with straighter walls and fewer missed inner roofline detections. It can predict buildings with common walls thanks to skeleton graph computation. To summarize, the proposed method is suitable for urban applications and has the potential to be improved further.

In recent years, intelligent identification and prediction of ore deposits based on deep learning algorithm and image processing technology has gradually become one of the research frontiers in the field of geological and metallogenic prediction. However, this method also has many problems that need to be solved. For example, (1) very few trainable image samples containing mineral point labels; (2) The geological image features are small and irregular, and the image similarity is high; (3) It is difficult to calculate the influence of different geological prospecting factors on ore mineralization. Based on this, this paper constructs a deep learning network model Multiscale Feature Attention Framework (MFAF) based on geo-image data. The results show that MFCA-Net module in MFAF model can solve the problem of scarce mine label images to a certain extent. In addition, the channel attention mechanism SE-Net module can quantify the difference of influence of different source factors on mineralization. The prediction map is obtained by applying the MFAF model in the study of deposit identification and prediction in the research area of the southern section of Qin-hang metallogenic Belt. In the forecast map, many regions have high metallogenic potential. Through observation, it is found that the above predicted area covers 100% of the known ore deposits at present, which has a good prediction effect. Later, we can find out the specific reasons through field verification. The multi-scale feature fusion and attention mechanism MFA Framework in this paper can provide a new way of thinking for geologists in mineral exploration. The research of this paper also provides resource guarantee and technical support for the sustainable exploitation of mineral resources and the sustainable growth of society and economy.

The unique characteristics of frescoes on overseas Chinese buildings can attest to the integration and historical background of Chinese and Western cultures. Reasonable analysis and preservation of overseas Chinese frescoes can provide sustainable development for culture and history. This research adopts the image analysis technology based on artificial intelligence, and proposes a ResNet-34 model and method integrating transfer learning. This deep learning model can identify and classify the source of the frescoes of the emigrants, and can effectively deal with the problems such as the small number of fresco images on the emigrants' buildings, poor quality, difficulty in feature extraction, and similar pattern text and style. The experimental results show that the training process of the model proposed in this article is stable. On the constructed Jiangmen and Haikou fresco JHD datasets, the final accuracy is 98.41%, and the recall rate is 98.53%. The above evaluation indicators are superior to classic models such as AlexNet, GoogLeNet, and VGGNet. It can be seen that the model in this article has strong generalization ability and is not prone to overfitting. It can effectively identify and classify the cultural connotations and regions of frescoes.

Medical brain image analysis is a necessary step in the Computers Assisted /Aided Diagnosis (CAD) systems. Advancements in both hardware and software in the past few years have led to improved segmentation and classification of various diseases. In the present work, we review the published literature on systems and algorithms that allow for classification, identification, and detection of White Matter Hyperintensities (WMHs) of brain MRI images specifically in cases of ischemic stroke and demyelinating diseases. For the selection criteria, we used the bibliometric networks. Out of a total of 140 documents we selected 38 articles that deal with the main objectives of this study. Based on the analysis and discussion of the revised documents, there is constant growth in the research and proposal of new models of deep learning to achieve the highest accuracy and reliability of the segmentation of ischemic and demyelinating lesions. Models with indicators (Dice Score, DSC: 0.99) were found, however with little practical application due to the uses of small datasets and lack of reproducibility. Therefore, the main conclusion is to establish multidisciplinary research groups to overcome the gap between CAD developments and their complete utilization in the clinical environment.

Prediction of protein-ligand binding affinities is crucial for computational drug discovery. A number of deep learning approaches have been developed in recent years to improve the accuracy of such affinity prediction. While the predicting power of these systems have advanced to some degrees depending on the dataset used for model training and testing, the effects of the quality and quantity of the underlying data have not been thoroughly examined. In this study, we employed erroneous datasets and data subsets of different sizes, created from one of the largest databases of experimental binding affinities, to train and evaluate a deep learning system based on convolutional neural networks. Our results show that data quality and quantity do have significant impacts on the prediction performance of trained models. Depending on the variations in data quality and quantity, the performance discrepancies could be comparable to or even larger than those observed among different deep learning approaches. In particular, the presence of proteins during model training leads to a dramatic increase in prediction accuracy. This implies that continued accumulation of high-quality affinity data, especially for new protein targets, is indispensable for improving deep learning models to better predict protein-ligand binding affinities.

Ever-increasing need for improving the livability of a city and improve outcomes for its residents, over the last decade, the adoption of technology to develop urbanised societies around the world has given rise to the need for developing smart cities. The speed at which the world population is growing, the use of Internet of Things in smart cities have really advanced the quality of life. One significant area of concern within the smart city framework is waste management. If the waste within a city is not adequately managed, then it leads to issues in the health of the citizens. Additionally, the waste management has such a high impact on the environmental footprint, hence the need to have a smart way of managing waste is of critical importance. Through our research, we analyse the challenges of waste management within a city to understand the impact of the problem on to the citizens and overall city operations. We then investigate ways in which we can solve these problems using the emerging technologies, such as the Internet of Things, to collect valuable data of large volumes arriving at an astronomical rate, then apply multi-agent deep reinforcement learning algorithms to harness the power of big data to extract meaningful information and actionable insights. We ingest data generated by our Internet of Things into our algorithm for three main purposes including providing the notifications to an external system, for example, a map navigation engine out of the scope for this project but a future extension for route optimisation and waste vehicle tracking; extracting and reporting the actionable insights from the underlying data; and consuming the extracted data for predictive forecasting to draw out the unknown patterns of waste fill levels within various geographical locations and again send out triggers and notification to external systems for example a waste collection authority who can efficiently schedule the waste collection vehicles and optimise the route. To achieve the above mentioned outcomes, we propose a framework that is agnostic of the hardware that it connects to and can effectively interface with a wide variety of hardware keeping a level of abstraction in the architecture.

Classifying network flow subsequences in an Industrial IoT gateway is an effective method for identifying Industrial IoT faults. This paper proposes a subsequence classification algorithm SSGBUL-IKNN based on unsupervised learning encoding to address the problem of low classification accuracy. Firstly, we built a network module based on unsupervised learning to encode the flow sequence of the Industrial IoT gateway. In the encoding module, a subsequence calibration algorithm controls the increasing error during the coding process. Then, the specific fault type is obtained through an improved KNN classification algorithm based on the indication distance source from the integrated coding sequences converted by multi-dimensional network flow sequences. The test result on three Industrial IoT datasets of a sewage treatment plant shows that the accuracy of SSGBUL-IKNN in this paper exceeds 92%, significantly higher than the algorithm DTW and TSF. It reaches the classification requirements of Industrial IoT fault diagnosis.

Diabetic retinopathy (DR) is the primary factor leading to vision impairment and blindness in diabetic people. Uncontrolled diabetes can damage the retinal blood vessels. Initial detection and prompt medical intervention are vital in preventing progressive vision impairment. Today's growing medical field poses more significant workload and diagnostic demands on medical professionals. In the proposed study, a convolutional neural network (CNN) is employed to detect the stages of DR. This research is crucial for studying DR because of its innovative methodology incorporating two different public datasets. This strategy enhances the model's capacity to generalize unseen DR images, as each dataset encompasses unique demographics and clinical circumstances. The network can learn and capture complicated hierarchical image features with asymmetric weights. Each image undergoes preprocessing using contrast-limited adaptive histogram equalization (CLAHE) and the discrete wavelet transform (DWT). The model is trained and validated using the combined datasets of Dataset for Diabetic Retinopathy (DDR) and the Asia Pacific Tele-Ophthalmology Society (APTOS). The CNN model is tuned with different learning rates and optimizers. The CNN model with Adam optimizer achieved 70% accuracy and an area under curve (AUC) score of 0.82. The recommended study results may reduce diabetes-related vision impairment by early DR severity identification.

The COVID-19 pandemic has wreaked havoc on the whole world, taking over half a million lives and capsizing the world economy in unprecedented magnitudes. With the world scampering for a vaccine, early detection and containment is the only redress. Existing diagnostic technologies with high accuracy like RT-PCRs are expensive and sophisticated, requiring skilled individuals for specimen collection and screening, resulting in lower outreach. So, methods excluding direct human intervention are much sought after, and artificial intelligence-driven automated diagnosis, especially with radiography images, captured the researchers' interest. This survey marks a detailed inspection of the deep-learning-based automated detection of COVID-19 works done to date, a comparison of the available datasets, methodical challenges like imbalanced datasets, and others, along with probable solutions with different pre-processing methods, and scopes of future exploration in this arena. We also benchmarked the performance of 315 deep models in diagnosing COVID-19, Normal, and Pneumonia from x-ray images of a custom dataset created from four others. The dataset is publicly available at https://github.com/rgbnihal2/COVID-19-X-ray-Dataset. Our results show that DenseNet201 model with Quadratic SVM classifier performs the best (accuracy: 98.16%, sensitivity: 98.93%, specificity: 98.77%) and maintains high accuracies in other similar architectures as well. This proves that even though radiography images might not be conclusive for radiologists, but it is so for deep learning algorithms for detecting COVID-19. We hope this extensive review will provide a comprehensive guideline for researchers in this field.

Despite the importance of few-shot learning, the lack of labeled training data in the real world, makes it extremely challenging for existing machine learning methods as this limited data set does not represent the data variance well. In this research, we suggest employing a generative approach using variational autoencoders (VAEs), which can be used specifically to optimize few-shot learning tasks by generating new samples with more intra-class variations. The purpose of our research is to increase the size of the training data set using various methods to improve the accuracy and robustness of the few-shot face recognition. Specifically, we employ the VAE generator to increase the size of the training data set, including the basic and the novel sets while utilizing transfer learning as the backend. Based on extensive experimental research, we analyze various data augmentation methods to observe how each method affects the accuracy of face recognition. We conclude that the face generation method we proposed can effectively improve the recognition accuracy rate to 96.47% using both the base and the novel sets.

Classical Arabic poetry is a form of Arabic literature, and almost every piece of Arabic poetry follows one of the 16 meters. The rhythmic framework of the poems or the verses is called meters. The deep learning (DL) model was implemented using TensorFlow in this work using a large dataset of Arabic poetry. The character level encoding was used to convert text to integers for classifying the full-verse and half-verse data. The study evaluates the data without removing diacritics from the text. The train-test-split method with a 70-15-15 split was employed for classification. 15% of the total data was considered unseen test data for all models. The work was conducted with multiple deep learning models, including the Long Short-Term Memory (LSTM), Gated Recurrent Units (GRU), and Bidirectional LSTM (Bi-LSTM). The Bi-LSTM model shows the best accuracy of all the models specified, with 97.53% for full-verse and 95.23% for half-verse without removing diacritic data.

The used car market has a high global economic importance, with more than 35 million cars sold yearly. Accurately predicting prices is a crucial task for both buyers and sellers to facilitate informed decisions in terms of opportunities or potential problems. Although various Machine Learning techniques have been applied to create robust prediction models, a comprehensive approach has yet to be studied. This research introduces two datasets from different markets, one with over 300,000 entries from Germany to serve as a training base for deep prediction models and a second dataset from Romania containing more than 15,000 car quotes used mainly to observe local traits. As such, we include extensive cross-market analyses by comparing the emerging Romanian market versus one of the world’s largest and most developed car markets, Germany. Our study uses several neural network architectures that capture complex relationships between car model features, individual add-ons, and visual features to predict used car prices accurately. Our models achieved a high R2 score exceeding 0.95 on both datasets, indicating their effectiveness in estimating used car prices. Moreover, we experimented with advanced convolutional architectures to predict car prices based solely on visual features extracted from car images. This approach exhibited transfer-learning capabilities, leading to improved prediction accuracy, especially since the Romanian training dataset is limited. Our experiments highlight the most important factors influencing the price, while our findings have practical implications for buyers and sellers in assessing the value of vehicles. At the same time, the insights gained from this study enable informed decision-making and provide valuable guidance in the used car market.

The intracranial pressure (ICP) signal, as monitored on patients in intensive care units, contains pulses of cardiac origin where P1 and P2 subpeaks can often be observed. When calculable, the ratio of their relative amplitudes is an indicator of the patient’s cerebral compliance. This characterization is particularly informative on the overall state of the cerebrospinal system. We developed a recurrent neural network-based framework for P2/P1 ratio computation that only takes a raw ICP signal as an input. Two tasks are performed, namely pulse classification and subpeak designation. Pulse classification was achieved with an area under the curve of 0.90 on a 4,344-pulse testing dataset, while the peak designation algorithm identified pulses with a P2/P1 ratio > 1 with a 97.3% accuracy. Although it still needs to be evaluated on a larger number of labeled recordings, our automated P2/P1 ratio calculation framework appears to be a promising tool for improving bedside monitoring devices.

Transcutaneous injection laryngoplasty is a well-known procedure for treating paralyzed vocal fold by injecting augmentation material into it. Hence, vocal fold localization plays a vital role in the preoperative planning as the fold location is required to determine the optimal injection route. In this communication, we propose a mirror environment based reinforcement learning (RL) algorithm for localizing the right and left vocal folds in preoperative neck CT. RL-based methods commonly showed noteworthy outcome in general anatomic landmark localization problem in the recent years. However, such methods suggest training individual agent for localizing each fold, though the right and left vocal folds are located in close proximity and have high feature-similarity. Utilizing the lateral symmetry between the right and left vocal folds, the proposed mirror environment allows for a single agent for localizing both the folds by treating the left fold as a flipped version of the right fold. Thus, localization of both folds can be trained using a single training session which utilizes the inter-fold correlation and avoids redundant feature learning. Experiment with 120 CT volumes showed improved localization performance and training efficiency of the proposed method compared with the standard RL method.

Aiming at the problem of model error and tracking dependence in the process of intelligent vehicle motion planning, an intelligent vehicle model transfer trajectory planning method based on deep reinforcement learning is proposed, which obtain an effective control action sequence directly. Firstly, an abstract model of the real environment is extracted. On this basis, Deep Deterministic Policy Gradient (DDPG) and vehicle dynamic model are adopted to jointly train a reinforcement learning model, and to decide the optimal intelligent driving maneuver. Secondly, the actual scene is transferred to equivalent virtual abstract scene by transfer model, furthermore, the control action and trajectory sequences are calculated according to trained deep reinforcement learning model. Thirdly, the optimal trajectory sequence is selected according to evaluation function in the real environment. Finally, the results demonstrate that the proposed method can deal with the problem of intelligent vehicle trajectory planning for continuous input and continuous output. The model transfer method improves the model generalization performance. Compared with the traditional trajectory planning, the proposed method output continuous rotation angle control sequence, meanwhile, the lateral control error is also reduced.

In tunnel excavation with boring machines, the tunnel face is supported to avoid collapse and minimise settlement. This article proposes the use of reinforcement learning, specifically the Deep Q-Network algorithm, to predict the face support pressure. The approach is tested both analytically and numerically. By using the soil properties ahead of the tunnel face and the overburden depth as the input, the algorithm is capable of predicting the optimal tunnel face support pressure, adapting to changes in geological and geometrical conditions.

Indoor localization (IL) is a significant topic of study with several practical applications. The area of IL has evolved greatly in recent years due to the introduction of numerous technologies such as WiFi, Bluetooth, cameras, and other sensors. Despite the growing interest in this field, there are numerous challenges and drawbacks that must be addressed to develop more accurate and sustainable systems for IL and its real-life applications. This review study gives an in-depth look into IL, covering the most promising artificial intelligence-based and hybrid strategies that have shown excellent potential in overcoming some of the limitations of classic methods. In addition, the paper investigates the significance of high-quality datasets and evaluation metrics in the design and assessment of IL algorithms. Furthermore, this overview study emphasizes the crucial role that machine learning techniques, such as deep learning and transfer learning, play in the advancement of IL. A focus on the importance of IL and the various technologies, methods, and techniques that are being used to improve it. Finally, The survey highlights the need for continued research and development to create more accurate and scalable techniques that can be applied across a range of industries, such as evacuation-egress routes, hazard-crime detection, smart occupancy-driven energy reduction and asset tracking and management.

Epilepsy is a chronic neurological disorder characterized by abnormal electrical activity in the brain, often leading to recurrent seizures. With approximately 50 million people worldwide affected by epilepsy, there is a pressing need for efficient and accurate methods to detect and diagnose seizures. Electroencephalogram (EEG) signals, which record the brain’s electrical activity, have emerged as a valuable tool in detecting epilepsy and other neurological disorders. Traditionally, the process of analyzing EEG signals for seizure detection has relied on manual inspection by experts, which is time-consuming, labor-intensive, and susceptible to human error. To address these limitations, researchers have turned to machine learning and deep learning techniques to automate the seizure detection process. In this work, we propose a novel method for epileptic seizure detection, leveraging the power of 1-D Convolutional layers in combination with Bidirectional Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) and Average pooling Layer as a unit. The performance of the proposed model is verified on the Bonn dataset. Our proposed model offers significant advancements over existing approaches for the detection of epileptic seizures, particularly for 2, 3, 4, and 5-class classification tasks. To assess the robustness and generalizability of our proposed architecture, we employ 5-fold cross-validation. By dividing the dataset into five subsets and iteratively training and testing the model on different combinations of these subsets, we obtain robust performance measures, including accuracy, sensitivity, and specificity. We demonstrate the effectiveness of the proposed architecture in accurately detecting epileptic seizures from EEG signals by using EEG signals of varying lengths. The results indicate its potential as a reliable and efficient tool for automated seizure detection, paving the way for improved diagnosis and management of epilepsy.

Coronavirus disease since December 2019 has significantly affected millions of people. Given the effect this disease has on the pulmonary systems of humans, there is a need for chest radiographic imaging (CXR) for monitoring the disease and preventing further deaths. Several studies have been shown that Deep Learning models can achieve promising results for COVID-19 diagnosis towards the CXR perspective. However, this research field is at an initial stage since there is a limited number of large CXR repositories regarding COVID-19. In this study, five deep learning models were analyzed and evaluated with the aim of identifying COVID-19 from chest X-Ray images. More specifically, we utilized the ResNet50, ResNet101, DenseNet121, DenseNet169 and InceptionV3 using Transfer Learning. All models were trained and validated on the largest publicly available repository for COVID-19 CXR images. Furthermore, they were evaluated on unknown data as well, that was not used for training or validation, authenticating their performance, and clarifying their usage in a medical scenario. All models achieved satisfactory performance where ResNet101 was the superior model achieving 96% in Precision, Recall, and Accuracy. Our outcomes show the potential of deep learning models on COVID-19 medical offering a promising way for the deeper understanding of COVID-19.

Accurate forecasts of ocean waves energy can not only reduce costs for investment but it is also essential for management and operation of electrical power. This paper presents an innovative approach based on the Long Short Term Memory (LSTM) to predict the power generation of an economical wave energy converter named “Searaser”. The data for analyzing is provided by collecting the experimental data from another study and the exerted data from numerical simulation of searaser. The simulation is done with Flow-3D software which has high capability in analyzing the fluid solid interactions. The lack of relation between wind speed and output power in previous studies needs to be investigated in this field. Therefore, in this study the wind speed and output power are related with a LSTM method. Moreover, it can be inferred that the LSTM Network is able to predict power in terms of height more accurately and faster than the numerical solution in a field of predicting. The network output figures show a great agreement and the root mean square is 0.49 in the mean value related to the accuracy of LSTM method. Furthermore, the mathematical relation between the generated power and wave height was introduced by curve fitting of the power function to the result of LSTM method.

Deep learning models are part of the family of artificial neural networks and, as such, it suffers of catastrophic interference when they learn sequentially. In addition, most of these models have a rigid architecture which prevents the incremental learning of new classes. To overcome these drawbacks, in this article we propose the Self-Improving Generative Artificial Neural Network (SIGANN), a type of end-to-end Deep Neural Network system which is able to ease the catastrophic forgetting problem when leaning new classes. In this method, we introduce a novelty detection model to automatically detect samples of new classes, moreover an adversarial auto-encoder is used to produce samples of previous classes. This system consists of three main modules: a classifier module implemented using a Deep Convolutional Neural Network, a generator module based on an adversarial autoencoder; and a novelty detection module, implemented using an OpenMax activation function. Using the EMNIST data set, the model was trained incrementally, starting with a small set of classes. The results of the simulation show that SIGANN is able to retain previous knowledge with a gradual forgetfulness for each learning sequence. Moreover, SIGANN can detect new classes that are hidden in the data and, therefore, proceed with incremental class learning.

Timely detection of surface damages on wind turbine blades is imperative for minimising downtime and avoiding possible catastrophic structural failures. With recent advances in drone technology, a large number of high-resolution images of wind turbines are routinely acquired and subsequently analysed by experts to identify imminent damages. Automated analysis of these inspection images with the help of machine learning algorithms can reduce the inspection cost, thereby reducing the overall maintenance cost arising from the manual labour involved. In this work, we develop a deep learning based automated damage suggestion system for subsequent analysis of drone inspection images. Experimental results demonstrate that the proposed approach could achieve almost human level precision in terms of suggested damage location and types on wind turbine blades. We further demonstrate that for relatively small training sets advanced data augmentation during deep learning training can better generalise the trained model providing a significant gain in precision.

In recent years, weeds is responsible for most of the agricultural yield losses. To deal with this threat, farmers resort to spraying pesticides throughout the field. Such method not only requires huge quantities of herbicides but impact environment and humans health. One way to reduce the cost and environmental impact is to allocate the right doses of herbicide at the right place and at the right time (Precision Agriculture). Nowadays, Unmanned Aerial Vehicle (UAV) is becoming an interesting acquisition system for weeds localization and management due to its ability to obtain the images of the entire agricultural field with a very high spatial resolution and at low cost. Despite the important advances in UAV acquisition systems, automatic weeds detection remains a challenging problem because of its strong similarity with the crops. Recently Deep Learning approach has shown impressive results in different complex classification problem. However, this approach needs a certain amount of training data but, creating large agricultural datasets with pixel-level annotations by expert is an extremely time consuming task. In this paper, we propose a novel fully automatic learning method using Convolutional Neuronal Networks (CNNs) with unsupervised training dataset collection for weeds detection from UAV images. The proposed method consists in three main phases. First we automatically detect the crop lines and using them to identify the interline weeds. In the second phase, interline weeds are used to constitute the training dataset. Finally, we performed CNNs on this dataset to build a model able to detect the crop and weeds in the images. The results obtained are comparable to the traditional supervised training data labeling. The accuracy gaps are 1.5% in the spinach field and 6% in the bean field.

This paper presents a deep learning-based multi-guidance line detection approach, enabling an autonomous four-wheeled agricultural robot to navigate strip farmland. The integration of Proportional-Integral-Derivative (PID) and Fuzzy Logic Control (FLC) systems optimizes the velocity and heading angle of robot, facilitating smooth navigation along identified guidance line. Enhance cornering maneuverability with real-time kinematics (RTK)-assisted Global Navigation Satellite System (GNSS) (RTK-GNSS) positioning. Additionally, the spraying system combined with deep learning can effectively identify weeds and crop nutrient deficiencies to achieve precise spraying. The guidance system prioritizes irrigation lines for navigation, with additional guidance from crop and furrow lines. Trigonometric analysis determines the angular deviation between the identified guidance line and the vertical line of the top view image. Experiments under diverse weather conditions demonstrate the stable navigation of robot at 12.5 cm/s, achieving up to 90% accuracy in weed and nutrient deficiency identification. The spraying accuracy for weeds and deficiencies averages 87% and 73%, respectively, underscoring the system's contribution to sustainable and precision horticulture practices.

Hypergravity accelerators are a type of large machinery used for gravity training or medical research. A failure of such large equipment can be a serious problem in terms of safety or costs. This paper proposes a prediction model that can proactively prevent failures that may occur in a hy-pergravity accelerator. The method proposed in this paper was to convert vibration signals to spectograms and perform classification training using a deep learning model. An experiment was conducted to evaluate the performance of the method proposed in this paper. A 4-channel accel-erometer was attached to the bearing housing, which is a rotor, and time-amplitude data were obtained from the measured values by sampling. The data were converted to a two-dimensional spectrogram, and classification training was performed using a deep learning model for four conditions of the equipment: Unbalance, Misalignment, Shaft Rubbing, and Normal. The ex-perimental results showed that the proposed method had a 99.5% F1-Score, which was up to 23% higher than the 76.25% for existing feature-based learning models.

Diseases of cereals caused by pathogenic fungi can significantly reduce crop yields. Many cultures are exposed to them. The disease is difficult to control on a large scale, thus one of the relevant approaches is the crop field monitoring, which helps to identify the disease at an early stage and take measures to prevent its spread. One of the effective control methods is disease identification based on the analysis of digital images with the possibility of obtaining them in field conditions using mobile devices. In this work, we propose a method for the recognition of five fungal diseases of wheat shoots (leaf rust, stem rust, yellow rust, powdery mildew, and septoria), both separately and in combination, with the possibility of identifying the stage of plant development. A set of 2414 images of wheat fungi diseases (WFD2020) was generated, for which expert labeling was performed by the type of disease. WFD2020 data are available freely at http://wfd.sysbio.ru/. In the process of creating this set, a method was applied to reduce the degeneracy of the training data based on the image hashing algorithm. The disease recognition algorithm is based on the convolutional neural network with the EfficientNet architecture. The best accuracy (0.942) was shown by a network with a training strategy based on augmentation and transfer of image styles. The recognition method was implemented by the authors as a bot on the Telegram platform, which allows assessing plants by lesions in the field conditions.

The precise determination of the State-of-Health (SOH) of lithium-ion batteries is critical in the domain of battery management systems. The proposed model in this research paper emulates any deep learning or machine learning model by utilizing a Look Up Table (LUT) memory to store all activation inputs and their corresponding outputs. The operation that follows the completion of training is referred to as the LUT memory preparation procedure. The lookup operation performed on this method simply substitutes for the inference process. This is achieved by discretizing the input data and features before binarizing them. The term for the aforementioned operation is the LUT inference method. The procedure was evaluated in this study using two distinct neural network architectures: a bidirectional long short-term memory (LSTM) architecture and a standard fully connected neural network (FCNN). It is anticipated that considerably greater efficiency and velocity will be achieved during the inference procedure when the pre-trained deep neural network architecture is inferred directly. The principal aim of this research is to construct a lookup table that effectively establishes correlations between the SOH of lithium-ion batteries and ensures a degree of imprecision that is tolerable. According to the results obtained from the NASA PCoE lithium-ion battery dataset, the proposed methodology exhibits performance that is largely comparable to that of the initial machine learning models. Utilising the error assessment metrics RMSE, MAE, and (MAPE), the accuracy of SOH prediction has been quantitatively evaluated. The indicators mentioned above demonstrate a significant degree of accuracy when predicting SOH.

A detailed literature review is performed in this study to address solutions for the full-body design and control of an aerial manipulator. Deep Reinforcement Learning methods are growing to be utilized recently to cope with various uncertainties. The pros and cons of these theories will be explained as well as introducing the advantages of Fuzzy Reinforcement Learning methods. State-of-the-Art, possible challenges, potential approaches, and a summary of desired precision devices are discussed in this study.

Following the occurrence of a typhoon, quick damage assessment related to residents can facilitate quick dispatch of house repair and disaster insurance works. Employing a deep learning method, this study used aerial photos of the Chiba prefecture obtained following the Typhoon Faxai in 2019 to automatically detect and evaluate the roof damage. This study comprised three parts: training deep learning model, detecting the roof damage using trained model, and classifying the level of roof damage. The detection object comprised roof outline, blue tarps, and roof completely destroyed. The roofs were divided into three categories: roof without damage, roof with blue tarps and roof completely destroyed. the F value obtained using the proposed method was higher than those obtained using other methods. In addition, it can be further divided into 5 levels from level 0 to 4. Finally, the spatial distribution of the roof damage was analyzed using ArcGIS tools. The proposed method is expected to provide certain reference for the real-time detection of the roof damage after the occurrence of a typhoon.

In this perspective article, we ponder topographical enhancements of artificial neural networks. In our recent paper in JMLR, we proved a quasi-equivalence between the network width and depth and also discussed the power of intra-links, which can be viewed as network height. In 1982, Hopfield published a network to model human associative memory, which contains many loops for dynamic evolution toward fixed points. Based on noising-denoising loops, diffusion models are recently developed to enable Bayesian modeling and inference with big data. Furthermore, we envision development of multi-AI-agent systems through “netware” engineering as a quantum leap of software engineering for emergent behaviors and autonomous AI at individual and population levels. We believe that the novel use of links and loops in space and time via multi-scale coupling would catalyze the next-generation neural networks.

Given the landscape of intricate and constantly evolving cyber threats, organizations demand refined strategies to deploy a Security Information and Event Management to support the management of a Cyber Security Operations Center. The dynamic nature of cyber threats complicates the efficient allocation of the location of network intrusion detection sensors, a critical component of a robust cybersecurity framework. Our research introduces an approach that integrates the precision of biomimetic optimization algorithms with the adaptability of Deep Q–Learning. By employing different biomimetic algorithms enhanced with deep learning, we aim to refine the deployment of sensors in network infrastructures, balancing the network security imperative against deployment costs. The results of computational tests demonstrate that the improved iterations leveraging Deep Q–Learning have outperformed their native counterparts. These findings underscore the importance of reinforcement learning, specifically through Deep Q–Learning, as a powerful tool to enhance the effectiveness of metaheuristics in addressing optimization challenges.

While machine learning (ML) has been quite successful in the field of structural health monitoring (SHM), its practical implementation has been limited. This is because ML model training requires data containing a variety of distinct damaged instances captured from a real structure and the experimental generation of such data is challenging. One way to tackle this issue is by generating training data through numerical simulations. However, simulated data cannot capture the bias and variance of experimental uncertainty. To overcome this problem, this work proposes a deep learning-based domain transformation method for transforming simulated data to the experimental domain. Use of this technique has been demonstrated for debonding location and size prediction of stiffened panels using a vibration-based method. The results are satisfactory for both, debonding location and size prediction. This domain transformation method can be used in any field where experimental data for training machine learning models is scarce.

Deep learning techniques are increasingly being recognized as effective image classifiers. Aside from their successful performance in past studies, the accuracies have varied in complex environments in comparison with the popularly applied machine learning classifiers. This study seeks to explore the feasibility for using a U-Net deep learning architecture to classify bi-temporal high resolution county scale aerial images to determine the spatial extent and changes of land cover classes that directly or indirectly impact tidal marsh. The image set used in the analysis is a collection of a 1-m resolution collection of National Agriculture Imagery Program (NAIP) tiles from 2009 and 2019 covering Beaufort County, South Carolina. The U-net CNN classification results were compared with two machine learning classifiers, the Random Trees (RT) and the Support Vector Machine (SVM). The results revealed a significant accuracy advantage in using the U-Net classifier (92.4%) as opposed to the SVM (81.6%) and RT (75.7%) classifiers for overall accuracy. From the perspective of a GIS analyst or coastal manager, the U-Net classifier is now an easily accessible nad powerful tool for mapping large areas. Change detection analysis indicated little areal change on marsh extent, though increased land development throughout the county has the potential to negatively impact the health of the marshes. Future work should explore applying the constructed U-Net classifier to coastal environments in large geographic areas, while also implementing other data sources (e.g., LIDAR, multispectral data) to enhance classification accuracy.

Clients are increasingly looking for fast and effective means to quickly and frequently survey and communicate the condition of their buildings so that essential repairs and maintenance work can be done in a proactive and timely manner before it becomes too dangerous and expensive. Traditional methods for this type of work commonly comprise of engaging building surveyors to undertake a condition assessment which involves a lengthy site inspection to produce a systematic recording of the physical condition of the building elements, including cost estimates of immediate and projected long-term costs of renewal, repair and maintenance of the building. Current asset condition assessment procedures are extensively time consuming, laborious, and expensive and pose health and safety threats to surveyors, particularly at height and roof levels which are difficult to access. We propose a method for automated detection and localisation of key building defects from images using deep learning and convolution neural networks. The proposed model is based on a pre-trained VGG-16 classifier with Class Activation Mapping (CAM) for object localisation. The model has proven to be robust and able to accurately detect and localise mould growth, stains, and paint deterioration defects arising from dampness in buildings. The approach is being developed with potentials to scale-up to support automated detection of defects and deterioration of buildings in real-time using mobile devices and drones.

In the extensive monitoring of maritime traffic, maritime management frequently encounters incomplete automatic identification system (AIS) data, particularly in critical static fields like vessel types. This shortfall presents substantial challenges in safety management, necessitating robust methods for data completion. Distinct from road traffic, where road width imposes constraints, maritime vessels of identical classifications often choose significantly divergent routes, deviating by several nautical miles. Sole dependence on trajectory clustering is insufficient. Addressing this issue, we segments the maritime area into spatiotemporal grids, transforming vessel paths into sequences of grid encodings. Utilizing natural language processing techniques, we integrate the Word2vec word embedding approach with our novel biLSTM self-attention chunk-max pooling net (biSAMNet) model, enhancing the classification of vessel trajectories. This method is crucial for determining static details like vessel types. Employing the Taiwan Strait as a case study and benchmarking against CNN, RNN, and methods based on the attention mechanism, our findings underscore our model’s superiority. The biSAMNet achieves an impressive vessel classification F1 score of 0.94, using only 5-dimensional word embeddings, highlighting the effectiveness of Word2vec pre-trained embedding layers. This research introduces a novel paradigm for processing vessel trajectory data, greatly enhancing the accuracy of filling in incomplete vessel type details.

For the generation of representative volume elements a statistical description of the relevant parameters is necessary. These parameters usually describe the geometric structure of a single grain. Commonly, parameters like area, aspect ratio and slope of the grain relative to the rolling direction are applied. However, usually simple distribution functions like log normal or gamma distribution are used. Yet, these do not take the interdependencies between the microstructural parameters into account. To fully describe any metallic microstructure though, these interdependencies between the singular parameters need to be accounted for. To accomplish this representation, a machine learning approach was applied in this study. By implementing a Wasserstein generative adversarial network, the distribution, as well as the interdependencies could accurately be described. A validation scheme was applied to verify the excellent match between microstructure input data and synthetically generated output data.

In our increasingly urbanized world, precise and up-to-date maps of human settlements are essential for sustainable urban development policies. The availability of open-access Sentinel-2 data from the Copernicus program presents an opportunity to create a comprehensive global map of human settlements, offering a detailed view of built areas on a large scale. This study estimates large-scale built-up fractions using encoder-decoder deep learning architectures like U-net, Res-U-net, and Attention-U-net in the large and complex urban area of Delhi, India. Openly available datasets like Open Street Map (OSM) and Microsoft building footprint datasets are used to derive built-up fractions at 10×10m resolution cells for over 34,000 km2. Our results show that Attention-U-net with the Huber loss function performs the best in different built-up densities (i.e., urban, semi-urban or rural) with an R2 score of 0.631, while Res-U-net and U-net obtained an R2 score of 0.623 and 0.612, respectively. The investigated networks significantly improve the accuracy over the latest Global Human Settlement Layer product (GHSL-S2), which uses a deep-CNN and reaches an R2 of 0.387 in our case study area. The result of this study yields a valuable spatial layer for examining the spatial distribution of human settlements across the entire spectrum from rural to urban areas.

With recent focus on deep neural network architectures for development of algorithms for computer-aided diagnosis (CAD), we provide a review of studies within the last 3 years (2015-2017) reported in selected top journals and conferences. 29 studies that met our inclusion criteria were reviewed to identify trends in this field and to inform future development. Studies have focused mostly on cancer-related diseases within internal medicine while diseases within gender-/age-focused fields like gynaecology/pediatrics have not received much focus. All reviewed studies employed image datasets, mostly sourced from publicly available databases (55.2%) and few based on data from human subjects (31%) and non-medical datasets (13.8%), while CNN architecture was employed in most (70%) of the studies. Confirmation of the effect of data manipulation on quality of output and adoption of multi-class rather than binary classification also require more focus. Future studies should leverage collaborations with medical experts to aid future with actual clinical testing with reporting based on some generally applicable index to enable comparison. Our next steps on plans for CAD development for osteoarthritis (OA), with plans to consider multi-class classification and comparison across deep learning approaches and unsupervised architectures were also highlighted.

Background: In today’s competitive market, predicting customer churn with high accuracy is crucial for enterprises to maintain growth and profitability. Traditional predictive models often lack in accuracy due to the complexity of customer behavior.Objective: This research aims to improve the accuracy of predicting customer churn by utilizing the Particle Swarm Optimization (PSO) algorithm for optimizing the hyperparameters of a composite deep learning model. The performance of this enhanced model is evaluated against traditional models such as LSRM_GRU, LSTM, GRU, and CNN_LSTM to demonstrate the effectiveness of PSO in hyperparameter tuning.Methods: A composite deep learning approach was employed, integrating various neural network architectures to leverage their strengths in modeling complex customer interactions. The PSO algorithm was used to optimize the model’s hyperparameters. Customer transaction and interaction data from different business operations served as the dataset for testing and analyzing the model’s performance. Evaluation metrics including accuracy, precision, recall, F1 score, and ROC AUC were utilized for a detailed comparison with established models.Findings: The PSO-enhanced composite deep learning model showed superior performance across all metrics, significantly outperforming the LSRM_GRU, LSTM, GRU, and CNN_LSTM models. Notably, improvements in ROC AUC and F1 score highlight the robustness and balanced precision-recall trade-off of the proposed model, demonstrating its effectiveness in identifying potential churners.Conclusion: The integration of PSO for hyperparameter optimization in composite deep learning models for customer churn prediction has proven to significantly enhance predictive accuracy and performance metrics over conventional models. This underscores the potential of evolutionary algorithms in improving deep learning applications. Future research should explore the scalability of this approach across various sectors and further refine predictive accuracy with evolving customer data.

In recent years, the viability of employing multi-agent reinforcement learning technology for adaptive traffic signal control has been extensively validated. However, owing to restricted communication among agents and the partial observability of the traffic environment, the process of mapping road network states to actions encounters numerous challenges. To address this problem, this paper proposes a multi-agent deep reinforcement learning model with an emphasis on communication content (CMARL). The model decouples the complex relationships between multi-signal agents through centralized training and decentralized execution. Specifically, we first pass the traffic state through an improved deep neural network to achieve the extraction of high-dimensional semantic information and the learning of the communication matrix. Then the agents selectively interact with each other based on the learned communication matrix and generate the final state features. Finally, the features are inputted to the QMIX network to achieve the final action selection. We compare the CMARL model with 6 other baseline algorithms in real traffic networks. The results show that CMARL can significantly reduce vehicle congestion, and run stably in various scenarios.

Semantic image segmentation has recently witnessed considerable progress by training deep convolutional neural networks (CNNs). The core issue of this technique is the limited capacity of CNNs to depict visual objects. Existing approaches tend to utilize approximate inference in a discrete domain or additional aides and do not have a global optimum guarantee. We propose the use of the multi-label manifold ranking (MR) method in solving the linear objective energy function in a continuous domain to delineate visual objects and solve these problems. We present a novel embedded single stream optimization method based on the manifold ranking (MR) model to avoid approximations without sacrificing expressive power. In addition, we propose a novel network, which we refer to as dual multi-scale manifold ranking (\textbf{DMSMR}) network, that combines the dilated, multi-scale strategies with the single stream MR optimization method in the deep learning architecture to further improve the performance. Experiments on high resolution images, including close-range and remote sensing datasets, demonstrate that the proposed approach can achieve competitive accuracy without additional aides in an end-to-end manner.

Deep learning (DL), a branch of machine learning (ML) and artificial intelligence (AI) is nowadays considered as a core technology of today's Fourth Industrial Revolution (4IR or Industry 4.0). Due to its learning capabilities from data, DL technology originated from artificial neural network (ANN), has become a hot topic in the context of computing, and is widely applied in various application areas like healthcare, visual recognition, cybersecurity, and many more. However, building an appropriate DL model is a challenging task, due to the dynamic nature and variations in real-world problems and data. Moreover, the lack of core understanding turns DL methods into black-box machines that hamper development at the standard level. This article presents a structured and comprehensive view on DL techniques including a taxonomy considering various types of real-world tasks like supervised or unsupervised. In our taxonomy, we take into account deep networks for supervised or discriminative learning, unsupervised or generative learning as well as hybrid learning and relevant others. We also summarize real-world application areas where deep learning techniques can be used. Finally, we point out ten potential aspects for future generation DL modeling with research directions. Overall, this article aims to draw a big picture on DL modeling that can be used as a reference guide for both academia and industry professionals.

Deep learning (DL), which is originated from an artificial neural network (ANN), is one of the major technologies of today's smart cybersecurity systems or policies to function in an intelligent manner. Popular deep learning techniques, such as Multi-layer Perceptron (MLP), Convolutional Neural Network (CNN or ConvNet), Recurrent Neural Network (RNN) or Long Short-Term Memory (LSTM), Self-organizing Map (SOM), Auto-Encoder (AE), Restricted Boltzmann Machine (RBM), Deep Belief Networks (DBN), Generative Adversarial Network (GAN), Deep Transfer Learning (DTL or Deep TL), Deep Reinforcement Learning (DRL or Deep RL), or their ensembles and hybrid approaches can be used to intelligently tackle the diverse cybersecurity issues. In this paper, we aim to present a comprehensive overview from the perspective of these neural networks and deep learning techniques according to today's diverse needs. We also discuss the applicability of these techniques in various cybersecurity tasks such as intrusion detection, identification of malware or botnets, phishing, predicting cyber-attacks, e.g. denial of service (DoS), fraud detection or cyber-anomalies, etc. Finally, we highlight several research issues and future directions within the scope of our study in the field. Overall, the ultimate goal of this paper is to serve as a reference point and guidelines for the academia and professionals in the cyber industries, especially from the deep learning point of view.

Several studies suggest that COVID-19 may be accompanied by symptoms such as a dry cough, muscle aches, sore throat, and mild to moderate respiratory illness. The symptoms of this disease indicate the fact that COVID-19 causes noticeable negative effects on the lungs. Therefore, considering the health status of the lungs using X-rays and CT scans of the chest can significantly help diagnose COVID-19 infection. Due to the fact that most of the methods that have been proposed to COVID-19 diagnose deal with the lengthy testing time and also might give more false positive and false negative results, this paper aims to review and implement artificial intelligence (AI) image-based diagnosis methods in order to detect coronavirus infection with zero or near to zero false positives and false negatives rates. Besides the already existing AI image-based medical diagnosis method for the other well-known disease, this study aims on finding the most accurate COVID-19 detection method among AI methods such as machine learning (ML) and artificial neural network (ANN), ensemble learning (EL) methods.

Vibration measurement serves as the basis for various engineering practices such as natural frequency or resonant frequency estimation. As image acquisition devices become cheaper and faster, vibration measurement and frequency estimation through image sequence analysis continue to receive increasing attention. In the conventional photogrammetry and optical methods of frequency measurement, vibration signals are first extracted before implementing the vibration frequency analysis algorithm. In this work, we demonstrated that frequency prediction can be achieved using a single feed-forward convolutional neural network. The proposed method is verified using a vibration signal generator and excitation system, and the result obtained was compared with that of an industrial contact vibrometer in a real application. Our experimental results demonstrate that the proposed method can achieve acceptable prediction accuracy even in unfavorable field conditions.

The intricate relationship between Gray-Level Co-occurrence Matrix (GLCM) metrics and machine learning model performance underscores the need for rigorous dataset evaluation and selection protocols to ensure the reliability and generalizability of classification outcomes. This study involved a thorough examination of selected publicly available plant diseases datasets, with an emphasis on how well they performed as measured by GLCM metrics. After first classifying the datasets according to their GLCM metrics, dataset_2 (D2) and dataset_5 (D5) were found, respectively, to be the best-performing dataset in all GLCM analyses. The same datasets were then used to train deep learning models, and their classification performances were assessed. A noteworthy association was observed between the results of training deep learning models and the performance ratings derived from GLCM studies. More specifically, dataset_2 (D2) performed best in both GLCM analysis and deep learning model performance, indicating a strong correlation between the accuracy of classification and the textural qualities that GLCM captured. In the context of plant disease identification, in particular, these results highlight the significance of clearly defined dataset selection criteria in deep learning applications. Scholars can improve the accuracy and dependability of deep learning models for diagnosing plant diseases by giving preference to datasets with favorable GLCM metrics. The research also emphasizes the importance of texture features being taken into account in addition to conventional image features, highlighting the necessity of transparency and rigor in dataset selection procedures.

Ransomware attacks have rapidly proliferated, inflicting severe financial damages on businesses and individuals. Machine learning approaches to automate ransomware detection have shown promise but grapple with challenges like limited training data. This study introduces a novel deep learning model for few-shot ransomware classification. The model employs entropy features derived directly from malware binaries coupled with a twin neural network architecture utilizing transfer learning. Tests on over 1000 samples across 11 families demonstrate a weighted F1-score of 85.8\%, surpassing existing methods. The approach mitigates biases in limited training data and preserves intricacies lost in image-based features. It exhibits precise classification capabilities even with sparse samples of new ransomware variants. The research highlights the potential of entropy-driven deep learning to equip defenses against emerging zero-day ransomware strains.

Purpose: To simplify the application of optical coherence tomography (OCT) for examination of subsurface morphology in the oral cavity and reduce barriers towards adoption of OCT as a biopsy guidance device. The aim of this work was to develop automated software tools for simplified analysis of the large volume of data collected during OCT. Methods: Imaging and corresponding histopathology were acquired in-clinic using a wide-field endoscopic OCT system. An annotated dataset (n = 294 images) from 60 patients (34 male, 26 female) was assembled to train four unique neural networks. A deep learning pipeline was built using convolutional and modified u-net models to detect the imaging field of view (network 1), detect artifacts (network 2), identify the tissue surface (network 3), and identify the presence and location of the basement membrane (network 4). Results: The area under the curve of the image and artifact detection networks was 1.00 and 0.94 respectively. The Dice similarity score for the surface and basement membrane segmentation networks was 0.98 and 0.83 respectively. Conclusion: Deep Learning (DL) techniques can identify the location and variations in the epithelial surface and basement membrane in OCT images of the oral mucosa. Segmentation results can be synthesized into accessible en face maps to allow easier visualization of changes.

The increasing growth in knowledge about functioning of the nervous system of mammals and humans, as well as the significant neuromorphic technologies development in recent decades, have led to the emergence of a large number of brain-computer interfaces and neuroprosthetics for the regenerative medicine tasks. Neurotechnologies have traditionally been developed for therapeutic purposes to help or replace motor, sensory or cognitive abilities damaged by injury or disease. They also have significant potential for memory enhancement. However, there are still no fully developed neurotechnologies and neural interfaces capable of restoring or expanding cognitive functions, in particular memory, in mammals or humans. In this regard, the search for new technologies in the field of restoration of cognitive functions is an urgent task of modern neurophysiology, neurotechnology and artificial intelligence. Hippocampus is an important brain structure connected to memory and information processing in the brain. The aim of this paper is to propose an approach based on deep neural networks for prediction of hippocampal signals in CA1 region based on received biological input in CA3 region. We compare results of prediction for two widely used deep architectures: reservoir computing (RC) and long short-term memory (LSTM) networks. Proposed study can be viewed as a first step in the complex task of the development of a neurohybrid chip, which allows one to restore memory functions in the damaged rodent hippocampus.

The use of visual signals in horticulture has attracted significant attention and encompassed a wide range of data types such as 2D images, videos, hyperspectral images, and 3D point clouds. These visual signals have proven to be valuable in developing cutting-edge computer vision systems for various applications in horticulture, enabling plant growth monitoring, pest and disease detection, quality and yield estimation, and automated harvesting. However, unlike other sectors, developing deep learning computer vision systems for horticulture encounters unique challenges due to the limited availability of high-quality training and evaluation datasets necessary for deep learning models. This paper investigates the current status of vision systems and available data in order to identify the high-quality data requirements specific to horticultural applications. We analyse the impact of the quality of visual signals on the information content and features that can be extracted from these signals. To address the identified data quality requirements, we explore the usage of a deep learning-based super-resolution model for generative quality enhancement of visual signals. Furthermore, we discuss how these can be applied to meet the growing requirements around data quality for learning-based vision systems. We also present a detailed analysis of the competitive quality generated by the proposed solution compared to cost-intensive hardware-based alternatives. This work aims to guide the development of efficient computer vision models in horticulture by overcoming existing data challenges and paving a pathway forward for contemporary data acquisition.

The fifth Generation (5G) mobile networks use millimeter Waves (mmWaves) to offer giga bit data rates. However, unlike microwaves, mmWave links are prone to user and topographic dynamics. They easily get blocked and end up forming irregular cell patterns for 5G. This in turn cause too early, too late, or wrong handoffs (HOs). To mitigate HO challenges, sustain connectivity and avert unnecessary HO, we propose a HO scheme based on Jump Markov Linear System (JMLS) and Deep Reinforcement Learning (DRL). JMLS is widely known to account for abrupt changes in system dynamics. DRL likewise emerges as an artificial intelligence technique for learning highly dimensional and time-varying behaviors. We combine the two techniques to account for time-varying, abrupt, and irregular changes in mmWave link behaviour by predicting likely deterioration patterns of target links. The prediction is optimized by meta training techniques that also reduces training sample size. Thus, the JMLS-DRL platform formulates intelligent and versatile HO policies for 5G. Results show our proposed prediction scheme about target link behavior post HO to be highly reliable. The scheme also averts unnecessary HOs thus ably supports longer dew time.

The ‘inverse problem’ of mass spectrometric molecular identification (‘given a mass spectrum, calculate/predict the 2D structure of the molecule whence it came’) is largely unsolved, and is especially acute in metabolomics where many small molecules remain unidentified. This is largely because the number of experimentally available electrospray mass spectra of small molecules is quite limited. However, the forward problem (‘calculate a small molecule’s likely fragmentation and hence at least some of its mass spectrum from its structure alone’) is much more tractable, because the strengths of different chemical bonds are roughly known. This kind of molecular identification problem may be cast as a language translation problem in which the source language is a list of high-resolution mass spectral peaks and the ‘translation’ a representation (for instance in SMILES) of the molecule. It is thus suitable for attack using the deep neural networks known as transformers. We here present MassGenie, a method that uses a transformer-based deep neural network, trained on ~6 million chemical structures with augmented SMILES encoding and their paired molecular fragments as generated in silico, explicitly including the protonated molecular ion. This architecture (containing some 400 million elements) is used to predict the structure of a molecule from the various fragments that may be expected to be observed when some of its bonds are broken. Despite being given essentially no detailed nor explicit rules about molecular fragmentation methods, isotope patterns, rearrangements, neutral losses, and the like, MassGenie learns the effective properties of the mass spectral fragment and valency space, and can generate candidate molecular structures that are very close or identical to those of the ‘true’ molecules. We also use VAE-Sim, a previously published variational autoencoder, to generate candidate molecules that are ‘similar’ to the top hit. In addition to using the ‘top hits’ directly, we can produce a rank order of these by ‘round-tripping’ candidate molecules and comparing them with the true molecules, where known. As a proof of principle, we confine ourselves to positive electrospray mass spectra from molecules with a molecular mass of 500Da or lower, including those in the last CASMI challenge (for which the results are known), getting 49/93 (53%) precisely correct. The transformer method, applied here for the first time to mass spectral interpretation, works extremely effectively both for mass spectra generated in silico and on experimentally obtained mass spectra from pure compounds. It seems to act as a Las Vegas algorithm, in that it either gives the correct answer or simply states that it cannot find one. The ability to create and to ‘learn’ millions of fragmentation patterns in silico, and therefrom generate candidate structures (that do not have to be in existing libraries) directly, thus opens up entirely the field of de novo small molecule structure prediction from experimental mass spectra.

The question of molecular similarity is core in cheminformatics, and is usually assessed via a pairwise comparison based on vectors of properties or molecular fingerprints. We recently exploited variational autoencoders to embed 6M molecules in a chemical space, such that their (Euclidean) distance within the latent space so formed could be assessed within the framework of the entire molecular set. However, the standard objective function used did not seek to manipulate the latent space so as to cluster the molecules based on any perceived similarity. Using a set of some 160,000 molecules of biological relevance, we here bring together three modern elements of deep learning to create a novel and disentangled latent space, viz transformers, contrastive learning, and an embedded autoencoder. The effective dimensionality of the latent space was varied such that clear separation of individual types of molecules could be observed within individual dimensions of the latent space. The capacity of the network was such that many dimensions were not populated at all. As before, we assessed the utility of the representation by comparing clozapine with its near neighbours, and did the same for various antibiotics related to flucloxacillin. Transformers, especially when as here coupled with contrastive learning, effectively provide one-shot learning, and lead to a successful and disentangled representation of molecular latent spaces that at once uses the entire training set in their construction while allowing ‘similar’ molecules to cluster together in an effective and interpretable way.

Measuring Marginal Reflex Distance-1 (MRD-1) is a crucial clinical tool used to evaluate the position of the eyelid margin in relation to the cornea. Traditionally, this assessment has been conducted manually by plastic surgeons, ophthalmologists, or trained technicians. However, with the advancements in Artificial Intelligence (AI) technology, there is a growing interest in the development of automated systems capable of accurately measuring MRD-1.In this context, we introduce novel MRD-1 measurement methods based on deep learning algorithms that can simultaneously capture images and compute the results. This prospective observational study involved 154 eyes of 77 patients aged over 18 years who visited Chungnam National University Hospital between September 1, 2023, and July 29, 2023. We collected four different MRD1 da-tasets from patients using three distinct measurement methods, each tailored to the individual patient. The mean MRD1 values, measured through the manual method using a penlight, the deep learning method, ImageJ analysis from RGB eye images, and ImageJ analysis from IR eye images in 56 eyes of 28 patients, were 2.64 ± 1.04 mm, 2.85 ± 1.07 mm, 2.78 ± 1.08 mm, and 3.07 ± 0.95 mm, respectively. Notably, the strongest agreement was observed between MRD1_deep learning (DL) and MRD1_IR (0.822, p < 0.01). In the Bland-Altman plot, the smallest difference was observed between MRD1_DL and MRD1_IR ImageJ, with a mean difference of 0.0611 and △LOA (limits of agreement) of 2.5162, which is the smallest among the other groups. In conclusion, this novel MRD1 measurement method, based on an IR camera and deep learning, demon-strates statistical significance and can be readily applied in clinical settings.

Deep Learning is a recent and important addition to the computational toolbox available for image reconstruction in fluorescence microscopy. We review state-of-the-art applications such as image restoration, super-resolution, and light-field imaging, and discuss how the latest Deep Learning research can be applied to other image reconstruction tasks such as structured illumination, spectral deconvolution, and sample stabilisation. Despite its successes, Deep Learning also poses significant challenges, has often misunderstood capabilities, and overlooked limits. We will address key questions, such as: What are the challenges in obtaining training data? Can we discover structures not present in the training data? And, what is the danger of inferring unsubstantiated image details?

The accurate classification of landfill waste diversion plays a critical role in efficient waste management practices. Traditional approaches, such as visual inspection, weighing and volume measurement, and manual sorting, have been widely used but suffer from subjectivity, scalability, and labour requirements. In contrast, machine learning approaches, particularly Convolutional Neural Networks (CNN), have emerged as powerful deep learning models for waste detection and classification. This paper analyses VGG-16, InceptionResNetV2, DenseNet121, Inception V3, and MobileNetV2 models to classify real-life waste when trained on pristine and unadulterated materials, versus samples collected at a landfill site. When training on DiversionNet, the unadulterated material dataset with labels required for landfill modelling, classification accuracy was limited to 49.69% in the real environment. Using real-world samples in the newly formed RealWaste dataset showed that practical applications for deep learning in waste classification is possible, with Inception V3 reaching 89.19% classification accuracy on the full spectrum of labels required for accurate modelling.

We present and benchmark FilterNet, a flexible deep learning architecture for time series classification tasks, such as activity recognition via multichannel sensor data. It adapts popular CNN and CNN-LSTM motifs which have excelled in activity recognition benchmarks, implementing them in a many-to-many architecture to markedly improve frame-by-frame accuracy, event segmentation accuracy, model size, and computational efficiency. We propose several model variants, evaluate them alongside other published models using the Opportunity benchmark dataset, demonstrate the effect of model ensembling and of altering key parameters, and quantify the quality of the models’ segmentation of discrete events. We also offer recommendations for use and suggest potential model extensions. FilterNet advances the state of the art in all measured accuracy and speed metrics on the benchmarked dataset, and it can be extensively customized for other applications.

Data-driven models (DDMs) are extensively used in environmental modeling but face challenges due to limited training data and potential results not adhering to physical laws. To address this challenge, this study developed a process-guided deep learning (PGDL) model, integrating a long short-term memory (LSTM) neural network and a process-based model (PBM), CE-QUAL-W2 (W2), to predict water temperature in a stratified reservoir. The PGDL included an energy constraint term from W2's thermal energy equilibrium into the cost function of the LSTM, besides the mean square error term. In PGDL, parameters were optimized by penalizing deviations from the energy law, ensuring adherence to physical constraints. Compared to LSTM, PGDL demonstrated enhanced satisfaction with the energy balance and superior performance in water temperature prediction. Even with less field data for training, PGDL outperformed both LSTM and calibrated W2 after pre-training with data generated using the uncalibrated W2. Therefore, integration of DDM with a PBM ensured physical consistency in water temperature prediction for complex stratified reservoirs with limited data. Moreover, pre-training the PGDL with PBM proved highly effective in mitigating bias and variance due to insufficient field measurement data.

Aluminum profile surface defects can greatly affect the performance, safety and reliability of products. Traditional human-based visual inspection is low accuracy and time consuming, and machine vision-based methods depend on hand-crafted features which need to be carefully designed and lack robustness. To recognize the multiple types of defects with various size on aluminum profiles, a multiscale defect detection network based on deep learning is proposed. Then, the network is trained and evaluated using aluminum profile surface defects images. Results show 84.6%, 48.5%, 96.9%, 97.9%, 96.9%, 42.5%, 47.2%, 100%, 100%, 43.3% average precision(AP) for the ten defect categories, respectively, with a mean AP of 75.8%, which illustrate the effectiveness of the network in aluminum profile surface defects detection. In addition, saliency maps also show the feasibility of the proposed network.

Accurate segmentation of the Left Ventricle (LV) in Echocardiogram (Echo) images is essential for cardiovascular analysis. Conventional techniques are labor-intensive and exhibit inter-observer variability. Deep learning has emerged as a powerful tool for automated medical image segmentation, offering advantages in speed and potentially superior accuracy. This study explores the efficacy of employing the YOLO (You Only Look Once) segmentation model for automated LV segmentation in Echo images. YOLO, a cutting-edge object detection model, achieves exceptional speed-accuracy balance through its well-designed architecture. It utilizes efficient dilated convolutional layers and bottleneck blocks for feature extraction, while incorporating innovations like path aggregation and spatial attention mechanisms. These attributes make YOLO a compelling candidate for adaptation to LV segmentation in Echo images. We posit that by fine-tuning a pre-trained YOLO based model on a well-annotated Echo image dataset, we can leverage the model's strengths in real-time processing and precise object localization to achieve robust LV segmentation. The proposed approach entails fine-tuning a pre-trained YOLO model on a rigorously labeled Echo image dataset. Model performance has been evaluated using established metrics such as mean Average Precision (mAP) at an Intersection over Union (IoU) threshold of 50% (mAP50) with 98.31%and across a range of IoU thresholds from 50% to 95% (mAP50:95) with 75.27%. Successful implementation of YOLO for LV segmentation has the potential to significantly expedite and standardize Echo image analysis. This advancement could translate to improved clinical decision-making and enhanced patient care.

Microgrids (MGs) have evolved as critical components of modern energy distribution networks, providing increased dependability, efficiency, and sustainability. Effective control strategies are essential for optimising MG operation and maintaining stability in the face of changing environmental and load conditions. Traditional rule-based control systems are extensively used due to their interpretability and simplicity. However, these strategies frequently lack the flexibility for complex and changing system dynamics. This paper provides a novel method called hybrid intelligent control for adaptive MG that integrates basic rule-based-control and deep learning techniques, including gated recurrent units (GRU), basic recurrent neural network (RNN) and long short-term memory (LSTM). The main target of this hybrid approach is to improve MG management performance by combining the strengths of basic rule-based systems and deep learning techniques. These deep learning techniques readily enhance and adapt control decisions based on historical data and domain-specific rules, leading to increasing system efficiency, stability, and resilience in adaptive MG. Our results show that the proposed method optimises MG operation, especially under demanding conditions such as variable renewable energy supply and unanticipated load fluctuations. This study investigates special RNN architectures and hyper-parameter optimization techniques with the aim of predicting power consumption and generation within the adaptive MG system. Our promising results show the highest-performing models indicating high accuracy and efficiency in power prediction. The finest-performing model accomplishes an R2 value close to 1, representing a strong correlation between predicted and actual power values.

Considering the significant advancements in autonomous vehicle technology, research in this field is of interest to researchers. To drive vehicles autonomously, controlling steer angle, gas hatch, and brakes need to be learned. The behavioral cloning method is used to imitate humans’ driving behavior. We created a dataset of driving in different routes and conditions and using the designed model, the output used for controlling the vehicle is obtained. In this paper, the Learning of Self-driving Vehicles Based on Real Driving Behavior Using Deep Neural Network Techniques (LSV-DNN) is proposed. We designed a convolutional network which uses the real driving data obtained through the vehicle’s camera and computer. The response of the driver is during driving is recorded in different situations and by converting the real driver’s driving video to images and transferring the data to an excel file, obstacle detection is carried out with the best accuracy and speed using the Yolo algorithm version 3. This way, the network learns the response of the driver to obstacles in different locations and the network is trained with the Yolo algorithm version 3 and the output of obstacle detection. Then, it outputs the steer angle and amount of brake, gas, and vehicle acceleration. The LSV-DNN is evaluated here via extensive simulations carried out in Python and TensorFlow environment. We evaluated the network error using the loss function. By comparing other methods which were conducted on the simulator’s data, we obtained good performance results for the designed network on the data from KITTI benchmark, the data collected using a private vehicle, and the data we collected.

Accurate cloud quantification is essential in climate change research. In this work, we construct an automated computer vision framework by synergistically incorporating deep neural networks and finite sector clustering to achieve robust whole sky image-based cloud classification, adaptive segmentation, and recognition under intricate illumination dynamics. A bespoke YOLOv8 architecture attains over 95% categorical precision across four archetypal cloud varieties curated from extensive annual observations (2020) at a Tibetan highland station. Tailor-made segmentation strategies adapted to distinct cloud configurations, allied with illumination-invariant image enhancement algorithms, effectively eliminate solar interference and substantially boost quantitative performance even in illumination-adverse analysis scenarios. Compared with the traditional threshold analysis method, the cloud quantification accuracy calculated within the framework of this paper is significantly improved. Collectively, the methodological innovations provide an advanced solution to markedly escalate cloud quantification precision levels imperative for climate change research, while offering a paradigm for cloud analytics transferable to various meteorological stations.

Accurate detection of forest restoration sites using deep learning is crucial for effective ecosystem management, optimizing reforestation efforts, and ensuring environmental sustainability. While various deep learning-based systems have been developed for this purpose, the influence of different neural network architectures on model performance during the identification of forest restoration sites remains underexplored. This study aimed to address this gap by exploring the optimal methodology for extracting and classifying candidate sites for forest restoration based on Convolutional Neural Networks (CNNs), specialized for image recognition. Four categories—arable land, road and barren, quarry, and forest—were defined as candidate sites for forest restoration. A dataset comprising 17,043 samples was divided into training (11,929) and validation (5,114) sets at a ratio of 7:3 for model learning. Model accuracy was evaluated using pixel accuracy (PA) and mean intersection over union (mean IoU). The Resnet152-U-Net model demonstrated a pixel accuracy of 95.2% and a mean IoU of 61.3%, indicating excellent performance in extracting candidate sites for forest restoration. This approach offers spatial and temporal advantages over conventional field surveys or aerial image-based assessments, potentially serving as valuable data for selecting future forest restoration sites.

With the rapid expansion of intelligent resource-constrained devices and high-speed communication technologies, Internet of Things (IoT) has earned a wide recognition as the primary standard for low-power lossy networks (LLNs). Nevertheless, IoT infrastructures are vulnerable to cyber-attacks due to the constraints in computation, storage, and communication capacity of the endpoint devices. From one side, the majority of newly developed cyber-attacks are formed by slightly mutating formerly established cyber-attacks to produce a new attack tending to be treated as a normal traffic through the IoT network. From the other side, the influence of coupling the deep learning techniques with cybersecurity field has become a recent inclination of many security applications due to their impressive performance. In this paper, we provide a comprehensive development of a new intelligent and autonomous deep learning-based detection and classification system for cyber-attacks in IoT communication networks leveraging the power of convolutional neural networks, abbreviated as (IoT-IDCS-CNN). The proposed IoT-IDCS-CNN makes use of the high-performance computing employing the robust CUDA based Nvidia GPUs and the parallel processing employing the high-speed I9-Cores based Intel CPUs. In particular, the proposed system is composed of three subsystems: Feature Engineering subsystem, Feature Learning subsystem and Traffic classification subsystem. All subsystems are developed, verified, integrated, and validated in this research. To evaluate the developed system, we employed the NSL-KDD dataset which includes all the key attacks in the IoT computing. The simulation results demonstrated more than 99.3% and 98.2% of cyber-attacks’ classification accuracy for the binary-class classifier (normal vs anomaly) and the multi-class classifier (five categories) respectively. The proposed system was validated using k-fold cross validation method and was evaluated using the confusion matrix parameters (i.e., TN, TP, FN, FP) along with other classification performance metrics including precision, recall, F1-score, and false alarm rate. The test and evaluation results of the IoT-IDCS-CNN system outperformed many recent machine-learning based IDCS systems in the same area of study.

Background and Objective: Segmentation of the femur in Dual Energy X-ray (DXA) images poses challenges due to reduced contrast, noise, bone shape variations, and inconsistent X-ray beam penetration. In this study, we investigate the relationship between noise and certain deep learning (DL) techniques for semantic segmentation of the femur to enhance segmentation and bone mineral density (BMD) accuracy by incorporating noise reduction methods into DL models. Methods: Convolutional neural network (CNN) based models were employed to segment femurs in DXA images and evaluate the effects of noise reduction filters on segmentation accuracy and its effect on BMD calculation. Various noise reduction techniques were integrated into DL-based models to enhance image quality before training. We assessed the performance of the fully convolutional neural network (FCNN) in comparison to noise reduction algorithms and manual segmentation methods. Results: Our study demonstrated that FCNN outperformed noise reduction algorithms in enhancing segmentation accuracy and enabling precise calculation of BMD. The FCNN-based segmentation approach achieved a segmentation accuracy of 98.84% and a correlation coefficient of 0.9928 for BMD measurements, indicating its effectiveness in the clinical diagnosis of osteoporosis. Conclusions: In conclusion, integrating noise reduction techniques into DL-based models significantly improves femur segmentation accuracy in DXA images. The FCNN model, in particular, shows promising results in enhancing BMD calculation and clinical diagnosis of osteoporosis. These findings highlight the potential of DL techniques in addressing segmentation challenges and improving diagnostic accuracy in medical imaging.

The quality prediction of quaternary structure models of a protein complex, in the absence of its true structure, is known as the Estimation of Model Accuracy (EMA). EMA is useful for ranking predicted protein complex structures and using them appropriately in biomedical research, such as protein-protein interaction studies, protein design, and drug discovery. With the advent of more accurate protein complex (multimer) prediction tools, such as AlphaFold2-Multimer and ESMFold, the estimation of the accuracy of protein complex structures has attracted increasing attention. Many deep learning methods have been developed to tackle this problem; however, there is a noticeable absence of a comprehensive overview of these methods to facilitate future development. Addressing this gap, we present a review of deep learning EMA methods for protein complex structures developed in the past several years, analyzing their methodologies and impacts. We also provide a prospective summary of some potential new developments for further improving the accuracy of the EMA methods.

Pig behavior is an integral part of health and welfare management, as pigs usually reflect their inner emotions through behavior change. The livestock environment plays a key role in pigs' health and wellbeing. A poor farm environment increases the toxic GHGs, which might deteriorate pigs' health and welfare. In this study a computer-vision-based automatic monitoring and tracking model was proposed to detect short-term pigs' physical activities in a compromised environment. The ventilators of the livestock barn were closed for an hour, three times in a day (07:00-08:00, 13:00-14:00, and 20:00-21:00) to create a compromised environment, which increases the GHGs level significantly. The corresponding pig activities were observed before, during, and after an hour of the treatment. Two widely used object detection models (YOLOv4 and Fast-er R-CNN) were trained and compared their performances in terms of pig localization and posture detection. The YOLOv4, which outperformed the Faster R-CNN model, coupled with a Deep-SORT tracking algorithm to detect and track the pig activities. The results showed that the pigs became more inactive with the increase in GHG concentration, reducing their standing and walking activities. Moreover, the pigs also shortened their sternal-lying posture increasing the lateral lying posture duration at higher GHG concentration. The high detection accuracy (mAP: 98.67%) and tracking accuracy (MOTA: 93.86% and MOTP: 82.41%) signify the models’ efficacy in monitoring and tracking pigs' physical activities non-invasively.

Aim: To determine whether an AI model and single sensor measuring acceleration and ECG could model cognitive and physical fatigue for a self-paced trail run. Methods: A field-based protocol of continuous fatigue repeated hourly induced physical (~45 minutes) and cognitive (~10 minutes) fatigue on one healthy participant. Physical load was a 3.8 km, 200 m vertical gain, trail run with acceleration and electrocardiogram (ECG) data collected using a single sensor. Cognitive load was a Multi Attribute Test Battery (MATB) and separate assessment battery including the Finger Tap Test (FTT), Stroop, Trail Making A and B, Spatial Memory, Paced Visual Serial Addition Test (PVSAT), and a vertical jump. A fatigue prediction model was implemented using a Convolutional Neural Network (CNN). Results: When the fatigue test battery results were compared for sensitivity to the protocol load, FTT right hand (R2 0.71) and Jump Height (R2 0.78) were the most sensitive while the other tests were less sensitive (R2 values Stroop 0.49, Trail Making A 0.29, Trail Making B 0.05, PVSAT 0.03, spatial memory 0.003). Best prediction results were achieved with a rolling average of 200 predictions (102.4 s), during set activity types, mean absolute error for ‘walk up’ (MAE200 12.5%) and range of absolute error for ‘run down‘ (RAE200 16.7%). Conclusion: We were able to measure cognitive and physical fatigue using a single wearable sensor during a practical field protocol including contextual factors in conjunction with a neural network model. This research has practical application to fatigue research in the field.

Interest in drone solutions in forestry applications is growing. Using drones, datasets can be captured flexibly and at high spatial and temporal resolutions when needed. In forestry applications, fundamental tasks include the detection of individual trees, tree species classification, bio-mass estimation, etc. Deep Neural Networks (DNN) have shown superior results when comparing with conventional machine learning methods such as Multi-Layer Perceptron (MLP) in cases of huge input data. The objective of this research was to investigate 3D convolutional neural networks (3D-CNN) to classify three major tree species in a boreal forest: pine, spruce, and birch. The proposed 3D-CNN models were employed to classify tree species in a test site in Finland. The classifiers were trained with a dataset of 3039 manually labelled trees. Then the accuracies were assessed by employing independent datasets of 803 records. To find the most efficient set of feature combination, we compare the performances of 3D-CNN models trained with hyperspectral (HS) channels, RGB channels, and canopy height model (CHM), separately and combined. It is demonstrated that the proposed 3D-CNN model with RGB and HS layers produces the highest classification accuracy. The producer accuracy of the best 3D-CNN classifier on the test dataset were 99.6%, 94.8%, and 97.4% for pines, spruces, and birches, respectively. The best 3D-CNN classifier produced ~5% better classification accuracy than the MLP with all layers. Our results suggest that the proposed method provides excellent classification results with acceptable performance metrics for HS datasets. Our results show that pine class was detectable in most layers. Spruce was most detectable in RGB data, while birch was most detectable in the HS layers. Furthermore, the RGB datasets provide acceptable results for many low-accuracy applications.

(1) Background: Evidence-based policymaking requires data about the local population's socioeconomic status (SES) at detailed geographical level, however, such information is often not available, or is too expensive to acquire. Researchers have proposed solutions to estimate SES indicators by analyzing Google Street View images, however, these methods are also resource-intensive, since they require large volumes of manually labeled training data. (2) Methods: We propose a methodology for automatically computing surrogate variables of SES indicators using street images of parked cars and deep multiple instance learning. Our approach does not require any manually created labels, apart from data already available by statistical authorities, while the entire pipeline for image acquisition, parked car detection, car classification, and surrogate variable computation is fully automated. The proposed surrogate variables are then used in linear regression models to estimate the target SES indicators. (3) Results: We implement and evaluate a model based on the proposed surrogate variable at 30 municipalities of varying SES in Greece. Our model has $R^2=0.76$ and a correlation coefficient of $0.874$ with the true unemployment rate, while it achieves a mean absolute percentage error of $0.089$ and mean absolute error of $1.87$ on a held-out test set. Similar results are also obtained for other socioeconomic indicators, related to education level and occupational prestige. (4) Conclusions: The proposed methodology can be used to estimate SES indicators at the local level automatically, using images of parked cars detected via Google Street View, without the need for any manual labeling effort.

Many problems in the fields of finance and actuarial science can be transformed into the problem of solving backward stochastic differential equations (BSDE) and partial differential equations (PDE) with jumps, which are often difficult to solve in high-dimensional cases. To solve this problem, this paper applies the deep learning algorithm to solve a class of high-dimensional nonlinear partial differential equations with jump terms and their corresponding backward stochastic differential equations (BSDE) with jump terms. Using the nonlinear Feynman-Kac formula, the problem of solving this kind of PDE is transformed into the problem of solving the corresponding backward stochastic differential equations with jump terms, and the numerical solution problem is turned into a stochastic control problem. At the same time, the gradient and jump process of the unknown solution are separately regarded as the strategy function and they are approximated respectively by using two multilayer neural networks as function approximators. Thus, deep learning-based method is used to overcome the “curse of dimensionality” caused by high-dimensional PDE with jump, and the numerical solution is obtained. In addition, this paper proposes a new optimization algorithm based on the existing neural network random optimization algorithm, and compares the results with the traditional optimization algorithm, and achieves good results. Finally, the proposed method is applied to three practical high-dimensional problems: Hamilton-Jacobi-Bellman equation, bond pricing under the jump Vasicek model and high-dimensional option pricing model with default risk. The proposed numerical method has obtained satisfactory accuracy and efficiency. The method has important application value and practical significance in investment decision-making, option pricing, insurance and other fields.

Unmanned Underwater Vehicles (UUVs) are pivotal in ocean exploration, research, and various industrial activities such as marine mining and offshore engineering. However, traditional methods of navigating these vehicles face significant challenges, mainly due to the inefficacy of electromagnetic waves in water, leading to signal loss. To address these limitations, researchers are increasingly turned learning, an artificial intelligence technique capable of learning from data-enhancing underwater navigation, mainly through visual Simultaneous Localization and Mapping (SLAM). This paper explores integrating deep learning methodologies and sensor technologies to revolutionize underwater navigation for UUVs. Proprioceptive sensors, along with exteroceptive sensors, are crucial in accurately measuring and comprehending the underwater environment. Additionally, the paper provides detailed insights into the processes of underwater SLAM, camera-based underwater positioning systems, sonar systems for underwater navigation, and the utilization of Lidar in underwater navigation. Furthermore, it delves into applying deep learning techniques in underwater SLAM, offering a comprehensive understanding of the innovative processes driving advancements in underwater vehicle navigation. By leveraging these advancements, this research aims to improve underwater navigation systems' precision, reliability, and adaptability, thereby unlocking new frontiers in ocean exploration and industrial applications for UUVs.

Pattern recognition techniques form the heart of most, if not all, incoherent linear shift-invariant systems. When an object is recorded using a camera, the object information gets sampled by the point spread function (PSF) of the system, replacing every object point with the PSF in the sensor. The PSF is a sharp Kronecker Delta-like function when the numerical aperture (NA) is large with no aberrations. When the NA is small, and the system has aberrations, the PSF appears blurred. In the above case, if the PSF is known, then the object information can be obtained by scanning the PSF over the recorded object intensity pattern and looking for pattern matching conditions through a mathematical process called correlation. In this study, a recently developed deconvolution method, the Lucy-Richardson-Rosen algorithm (LR2A), has been implemented to computationally refocus images recorded in the presence of spatio-spectral aberrations. The performance of LR2A was compared against the Lucy-Richardson algorithm and non-linear reconstruction. LR2A exhibits a superior deconvolution capability even in extreme cases of spatio-spectral aberrations and blur. Experimental results of deblurring a picture captured using high-resolution smartphone cameras are presented. LR2A was implemented to significantly improve the performances of the widely used deep convolutional neural networks for image classification.

We have developed deep learning models for automatic identification of the maternal heart rate (MHR) and, more generally, false signals (FSs) on fetal heart rate (FHR) recordings. The models can be used to preprocess FHR data prior to automated analysis or as a clinical alert system to assist the practitioner. Three models were developed and used to detect (i) FSs on the MHR channel (the FSMHR model), (ii) the MHR and FSs on the Doppler FHR sensor (the FSDop model), and (iii) FSs on the scalp ECG channel (the FSScalp model). The FSDop model was the most useful because FSs are far more frequent on the Doppler FHR channel. All three models were based on a multilayer, symmetric, GRU and were trained on data recorded during the first and second stages of delivery. The FSMHR and FSDop models were also trained on antepartum recordings. The training dataset contained 1030 expert-annotated periods (mean duration: 36 min) from 635 recordings. In an initial evaluation of routine clinical practice, 30 fully annotated recordings for each sensor type (mean duration: 5 h for MHR and Doppler sensors, and 3 h for the scalp ECG sensor) were analyzed. The sensitivity, positive predictive value (PPV) and accuracy were respectively 62.20%, 87.1% and 99.90% for the FSMHR model, 93.1%, 95.6% and 99.68% for the FSDop model, and 44.6%, 87.2% and 99.93% for the FSScalp model. We built a second test dataset with a more solid ground truth by selecting 45 periods (lasting 20 min, on average) on which the Doppler FHR and scalp ECG signals were recorded simultaneously. Using scalp ECG data, the experts estimated the true FHR value more reliably and thus annotated the Doppler FHR channel more precisely. The models achieved a sensitivity of 53.3%, a PPV of 62.4%, and an accuracy of 97.29%. In comparison, two experts (blinded to the scalp ECG data) respectively achieved a sensitivity of 15.7%, a PPV of 74.3%, and an accuracy of 96.91% and a sensitivity of 60.7%, a PPV of 83.5% and an accuracy of 98.24%. Hence, the models performed at expert level (better than one expert and worse than the other), although a well-trained expert with good knowledge of FSs could probably do better in some cases. The models and datasets have been included in the Fetal Heart Rate Morphological Analysis open-source MATLAB toolbox and can be used freely for research purposes.

Antimicrobial peptides (AMPs) are a promising alternative to antibiotics to combat drug resistance in pathogenic bacteria. However, the development of AMPs with high potency and specificity remains a challenge, and new tools to evaluate antimicrobial activity are needed to accelerate the discovery process. As a step toward direct prediction of the experimental minimum inhibitory concentration (MIC) of AMPs, we proposed MBC-Attention, a combination of a multi-branch CNN architecture and attention mechanism. Using a curated dataset of 3929 AMP against Escherichia coli, the optimal MBC-Attention model achieved an average Pearson correlation coefficient of 0.775 and an RMSE of 0.533 (log μM) in three independent tests of 393 sequences each. This results in a 5–12% improvement in PCC and 7–13% improvement in RMSE compared with RF and SVM models. Ablation studies confirmed that both attention mechanisms contributed to performance improvement.

Commercial buildings situated in hot and humid tropical climates rely significantly on cooling systems to maintain optimal occupant comfort. A well-accurate day-ahead load profile prediction plays a pivotal role in planning the energy requirements of cooling systems. Despite the pressing need for effective day-ahead cooling load predictions, current methodologies have not fully harnessed the potential of advanced deep-learning techniques. This research paper aims to address this gap by investigating the application of innovative deep learning models in day-ahead hourly cooling load prediction for commercial buildings situated in tropical climates. A range of deep learning techniques, including Deep Neural Network, Convolutional Neural Networks, Recurrent Neural Networks, and Long Short-Term Memory networks are employed to enhance prediction accuracy. Furthermore, these individual deep learning techniques are synergistically integrated to create hybrid models, such as CNN-LSTM and Sequence-to-Sequence models. An extensive comparative analysis is conducted to identify the most effective hybrid model. Sequence-to-sequence model provided better performance compared to the other single and hybrid models. Experiments are conducted to choose the time horizons from the past that can serve as input to the models. Further, the influence of various categories of input parameters on prediction performance has been assessed. Historical cooling load, calendar features and outdoor weather parameters are found in decreasing order of influence on prediction accuracy. This research focuses on buildings located in Singapore and presents a comprehensive case study to validate the proposed models and methodologies.

The health and stability of trees are essential information for the safety of people and property in urban greenery, parks or along roads. The stability of the trees is linked to root stability but essentially also to trunk decay. Currently used internal tree stem decay assessment methods, such as tomography or penetrometry, are reliable but usually time-consuming and unsuitable for large-scale surveys. Therefore, a new method based on close-range remote sensed data, specifically close-range photogrammetry and iPhone LiDAR, was tested to detect decayed standing tree trunks automatically. The proposed study used the PointNet deep learning algorithm for 3D data classification. It was verified in three different datasets consisting of pure coniferous trees, pure deciduous trees or mixed data to eliminate the influence of the detectable symptoms for each group and species itself. The mean achieved validation accuracies of the models were 65.5 % for Coniferous trees, 58.4 % for Deciduous trees and 57.7 % for Mixed data classification. The accuracies indicate promising data, which can be either used by practitioners for preliminary surveys or for other researchers to acquire more input data and create more robust classification models.

Urinary cytology is a useful, essential diagnostic method in routine urological clinical practice. Liquid-based cytology (LBC) for urothelial carcinoma screening is commonly used in the routine clinical cytodiagnosis because of its high cell collection rate. Since conventional screening processes by cytoscreeners and cytopathologists using microscopes is limited in terms of human resources, it is important to integrate new deep learning methods that can automatically and rapidly diagnose a large amount of specimens without delay. The goal of this study was to investigate the use of deep learning models for the classification of urine LBC whole-slide images (WSIs) into neoplastic and non-neoplastic (negative). We trained deep learning models using 786 WSIs by transfer learning, fully supervised, and weakly supervised learning approaches. We evaluated the trained models on two test sets (equal and clinical balance) with a combined total of 750 WSIs, achieving ROC-AUCs for WSI diagnosis in the range of 0.984-0.990 by the best model, demonstrating the promising potential use of our model for aiding urine cytodiagnostic processes.

Leader-follower autonomy (LFA) systems have so far only focused on vehicles following other vehicles. Though there have been several decades of research into this topic, there has not yet been any work on human-vehicle leader-follower systems. We present a system in which an autonomous vehicle — our ACTor 1 platform — can follow a human leader who controls the vehicle through gestures. We successfully developed a modular pipeline that uses artificial intelligence/deep learning to recognize hand-and-body gestures from a user in view of the vehicle’s camera and translate those gestures into physical action by the vehicle. We demonstrate our work using our ACTor 1 platform, a modified Polaris Gem 2 electric vehicle. This work has numerous applications, such as material transport in industrial contexts.

The advent of deep learning (DL) has revolutionized medical imaging, offering unprecedented avenues for accurate disease classification and diagnosis. DL models have shown remarkable promise for classifying brain tumor from Magnetic Resonance Imaging (MRI) scans. However, despite their impressive performance, the opaque nature of DL models poses challenges in understanding their decision-making processes, particularly crucial in medical contexts where interpretability is essential. This paper explores the intersection of medical image analysis and DL interpretability, aiming to elucidate the decision-making rationale of DL models in brain tumor classification. Leveraging state-of-the-art DL frameworks with transfer learning, we conduct a comprehensive evaluation encompassing both classification accuracy and interpretability. Using state-of-the-art DL frameworks with transfer learning, we conduct a thorough evaluation covering both classification accuracy and interpretability. We employ adaptive path-based techniques to understand the underlying decision-making mechanisms of these models. Grad-CAM and Grad-CAM++ highlight critical image regions where the tumors are located.

Accurate geometric modeling of blood vessels lumen given their 3D images is a basis for vessel quantification in support of cardiovascular diseases diagnosis, treatment and monitoring. Unlike other approaches, which assume circular or elliptical vessel cross-section, the proposed method employs parametric B-splines combined with the image formation system equations to accurately localize the highly curved lumen boundaries. No image segmentation is involved which might reduce the localization accuracy in effect of spatial discretization. It is shown that the model parameters can be reliably identified by a feedforward neural network which, driven by the cross-section images, predicts the parameter values many times faster than a reference least-squares (LS) model fitting algorithm. Two example applications: modeling the lower extremities artery-vein complexes visualized in steady-state contrast-enhanced magnetic resonance images (MRI) and the coronary arteries pictured in computed tomography angiograms (CTA), are presented and discussed to demonstrate the method potential. Besides application to medical diagnosis, blood-flow simulation and vessel-phantom design, it can serve as a tool for automated annotation of image datasets required to train machine-learning algorithms.

Industrial data scarcity is one of the largest factors holding back the widespread use of machine learning in manufacturing. To overcome this problem, the concept of transfer learning was developed and it achieved high attention in recent industrial research. Our paper focuses on the problem of time series segmentation and presents the first in-depth research about transfer learning for deep-learning based time series segmentation on the example of industrial end-of-line pump testing. In particularly, we investigate if the performance of deep learning models can be increased by pretraining the network with data from other domains. Three different scenarios are analyzed: source and target data being closely related, source and target data being distantly related, and source and target data being non-related. The results demonstrate that transfer learning can enhance the performance of time series segmentation models in respect to accuracy and training speed. The benefit is most clearly seen in scenarios where source and training data are closely related and the number of target training data samples is lowest. However, in the scenario of non-related datasets, cases of negative Transfer Learning were observed as well. Thus, the research emphasizes the potential, but also the challenges of industrial Transfer Learning.

Macula fovea detection is a crucial prerequisite towards screening and diagnosing macular diseases. Without early detection and proper treatment, any abnormality involving the macula may lead to blindness. However, with the ophthalmologist shortage and time-consuming artificial evaluation, neither accuracy nor effectiveness of the diagnose process could be guaranteed. In this project, we proposed a deep learning approach on ultra-widefield fundus (UWF) images for macula fovea detection. This study collected 2300 ultra-widefield fundus images from Shenzhen Aier Eye Hospital in China. Methods based on U-shape network (Unet) and Fully Convolutional Networks (FCN) are implemented on 1800 (before amplifying process) training fundus images, 400 (before amplifying process) validation images and 100 test images. Three professional ophthalmologists were invited to mark the fovea. A method from the anatomy perspective is investigated. This approach is derived from the spatial relationship between macula fovea and optic disc center in UWF. A set of parameters of this method is set based on the experience of ophthalmologists and verified to be effective. Results are measured by calculating the Euclidean distance between proposed approaches and the accurate grounded standard, which is detected by Ultra-widefield swept-source optical coherence tomograph (UWF-OCT) approach. Through a comparation of proposed methods, we conclude that, deep learning approach of Unet outperformed other methods on macula fovea detection tasks, by which outcomes obtained are comparable to grounded standard method.

Chest X-ray (CXR) is the first tool globally employed to detect cardiopulmonary pathologies. These acquisitions are highly affected by scattered photons due to the large field-of-view required. Scatter in CXRs introduces background in the images, which reduces their contrast. We developed three deep learning-based models to estimate and correct the scatter contribution to CXRs. We used a Monte Carlo (MC) ray-tracing model to simulate CXRs from human models obtained from CT scans using different configurations (depending on the availability of dual-energy acquisitions). The simulated CXRs contained the separated contribution of direct and scattered X-rays in the detector. These simulated datasets were then used as the reference for the supervised training of several NN. Three NN models (single and dual energy) were trained with the MultiResUNet architecture. The performance of the NN models was evaluated on CXRs obtained with the MC code from chest CTs of patients affected by COVID-19. The results showed that the NN models were able to estimate and correct the scatter contribution to CXRs with an error &lt;5%, being robust to variations in the simulation setup and improving the contrast in soft tissue. The single-energy model was tested with real CXRs, providing robust estimations of the scatter-corrected CXRs.

With metal oxide semiconductors being a promising candidate for accurate indoor air quality assessments, multiple drawbacks of the gas sensors prevent their widespread use. Examples include poor selectivity, instability over time, and sensor poisoning. Complex calibration methods and advanced operation modes can solve some of those drawbacks. However, this leads to long calibration times, which are unsuitable for mass production. In recent years, multiple attempts to solve calibration transfer have been made with the help of direct standardization, orthogonal signal correction, and many more methods. Besides those, a new promising approach is transfer learning from deep learning. This article will compare different calibration transfer methods, including direct standardization, piecewise direct standardization, transfer learning for deep learning models, and global model building. The machine learning methods to calibrate the initial models for calibration transfer are feature extraction, selection, and regression (established methods) and a custom convolutional neural network TCOCNN. It is shown that transfer learning can outperform the other calibration transfer methods regarding the root mean squared error, especially if the initial model is built with multiple sensors. It was possible to reduce the number of calibration samples by up to 99.3 % (from 10 days to approximately 2 hours) and still achieve an RMSE for acetone of around 18 ppb (15 ppb with extended individual calibration) if six different sensors were used for building the initial model. Furthermore, it was shown that the other calibration transfer methods (direct standardization and piecewise direct standardization) also work reasonably well for both machine learning approaches, primarily when multiple sensors are used for the initial model.

Deep learning has been applied in many areas that have had a significant impact on applications that improves real-life challenges. The success of deep learning in a wide range of areas is due in part to the use of activation functions, which are particularly effective at solving non-linear problems. Activation functions are a key focus for researchers in artificial intelligence who aim to improve the performance of neural networks. This article provides a comprehensive explanation and comparison of different activation functions, with a focus on the arc tangent and its variations specifically. The paper presents experimental results that show that variations of the arc tangent using irrational numbers such as pi, the golden ratio, and Euler’s number, as well as a self-arctan function, produce promising results. Since we experimented with promising activation functions on two different problems, and datasets, we reached a result that different irrationals work well for different problems. In other words, arctan ϕ gives the best results mostly for multiclass classification and arctan e gives the best results for time series prediction problems. The paper focuses on a multiclass classification problem applied to the Reuters Newswire dataset and a time-series prediction problem on Türkiye energy trade value to show the impacts of activation functions.

With the proliferating of video surveillance system deployment and related applications, real-time video analysis is very critical to achieve intelligent monitoring, autonomous driving, etc. It is non-trivial to achieve high accuracy and low latency video stream analysis through the traditional cloud computing. In this paper, we propose a non-orthogonal multiple access (NOMA) based edge real-time video analysis framework with one edge server (ES) and multiple user equipments (UEs). A cost minimization problem composed of delay, energy and accuracy is formulated to improve the QoE of UEs. In order to efficiently solve this problem, we propose the joint video frame resolution scaling, task offloading, and resource allocation algorithm based on the Deep Q-Learning Network (JVFRS-TO-RA-DQN), which effectively overcomes the sparsity of the single-layer reward function and accelerates the training convergence speed. JVFRS-TO-RA-DQN consists of two DQN networks to reduce the curse of dimension, which respectively select the offloading and resource allocation action, the resolution scaling action. Experimental results show that JVFRS-TO-RA-DQN can effectively reduce the cost of transmission and computation, and have better performance in convergence compared to other baseline.

The work presented in this paper makes multiple scientific contributions related to the investigation of the global fear associated with COVID-19 by performing a comprehensive analysis of a dataset comprising survey responses of participants from 40 countries. First, the results of subjectivity analysis of responses where participants indicated their biggest concern related to COVID-19 showed that the average subjectivity in responses by the age group of 41-50 decreased from April 2020 to June 2020, the average subjectivity in responses by the age group of 71-80 drastically increased from May 2020, and the age group of 11-20 indicated the least level of subjectivity in their responses between June 2020 to August 2020. Second, subjectivity analysis also revealed the percentage of highly opinionated, neutral opinionated, and least opinionated responses per age-group where the analyzed age groups were 11-20, 21-30, 31-40, 41-50, 51-60, 61-70, 71-80, and 81-90. For instance, the percentage of highly opinionated, neutral opinionated, and least opinionated responses by the age group of 11-20 were 17.92%, 16.24%, and 65.84%, respectively. Third, data analysis of responses from different age groups showed that the highest percentage of responses indicating that they were very worried about COVID-19 came from individuals in the age group of 21-30. Fourth, data analysis of the survey responses also revealed that in the context of taking precautions to prevent contracting COVID-19, the percentage of individuals in the age group of 31-40 taking precautions was higher as compared to the percentages of individuals from the age groups of 41-50, 51-60, 61-70, 71-80, and 81-90. Finally, a deep learning model was developed to detect if the survey respondents were seeing or planning to see a psychologist or psychiatrist for any mental health issues related to COVID-19. The deep learning model used the responses to multiple questions in the context of fear, preparedness, and response related to COVID-19 from the dataset and achieved an overall performance accuracy of 91.62% after 500 epochs.

One of the most common and important destructive attacks on the victim system is Advanced Persistent Threats (APT)-attack. The APT attacker can achieve his hostile goals by obtaining information and gaining financial benefits regarding the infrastructure of a network. One of the solutions to detect a secret APT attack is using network traffic. Due to the nature of the APT attack in terms of being on the network for a long time and the fact that the network may crash because of high traffic, it is difficult to detect this type of attack. Hence, in this study, machine learning methods such as C5.0 decision tree, Bayesian network and deep neural network are used for timely detection and classification of APT-attacks on the NSL-KDD data set. Moreover, 10-fold cross validation method is used to experiment these models. As a result, the accuracy (ACC) of the C5.0 decision tree, Bayesian network and 6-layer deep learning models is obtained as 95.64%, 88.37% and 98.85%, respectively, and also, in terms of the important criterion of the false positive rate (FPR), the FPR value for the C5.0 decision tree, Bayesian network and 6-layer deep learning models is obtained as 2.56, 10.47 and 1.13, respectively. Other criterions such as sensitivity, specificity, accuracy, false negative rate and F-measure are also investigated for the models, and the experimental results show that the deep learning model with automatic multi-layered extraction of features has the best performance for timely detection of an APT-attack comparing to other classification models.

In the current natural environment, due to the complexity of the background and the high similarity of the color between immature green tomato and plant, the occlusion of the key organs (flower and fruit) by the leaves and stems will lead to low recognition rate and poor generalization of the detection model. Therefore, an improved tomato organ detection method based on convolutional neural network has been proposed in this paper. Based on the original Faster R-CNN algorithm, Resnet-50 with residual blocks was used to replace the traditional vgg16 feature extraction network, and K-means clustering method was used to adjust more appropriate anchor size than manual setting to improve detection accuracy. A variety of data augmentation techniques were used to train the network. The test results showed that compared with the traditional Faster R-CNN model, the mean average precision (mAP) of the optimal model was improved from 85.2% to 90.7%, the memory requirement decreased from 546.9MB to 115.9 MB, and the average detection time was shortened to 0.073S/sheet. As the performance greatly improved, the training model can be transplanted to the embedded system, which lays a theoretical foundation for the development of precise targeting pesticide application system and automatic picking device.

Change detection with heterogeneous remote sensing images (Hete-CD) plays an important role in practical applications, especially when homogenous remote sensing images are unavailable. However, bitemporal heterogeneous remote sensing images (HRSIs) cannot compare directly to measure change magnitude, and many deep learning methods require large amounts of samples to train the module. Moreover, labeling many samples for land cover change detection with HRSIs is time-consuming and labor-intensive. Therefore, acquiring satisfactory performance of Hete-CD remains a challenge for deep learning networks with very small training samples. In this study, we promote a novel deep-learning framework for Hete-CD to obtain satisfactory performance while the initial samples are very small. We initially design a multiscale network with select kernel-attention module to focus on capturing different change targets with various sizes and shapes. Then, a simple yet effective non-parametric sample-enhanced algorithm based on the Pearson correlation coeffi-cient is promoted to explore potential samples around each initial sample. Finally, the proposed network and sample-enhanced algorithm are fused into one iterative framework to improve the change detection performance with very small samples. Experimental results conducted on four pairs of actual HRSIs indicated that the proposed framework can achieve competitive accuracies with very small samples for initialization when compared with some state-of-the-art methods. For example, the improvement is approximately 3.38% and 1.99% when compared with the selected traditional methods and deep learning methods, respectively.

Recently, the number of vehicles on the road, especially in urban centers, has increased dramatically due to the increasing trend of individuals towards urbanization. As a result, manual detection and recognition of vehicles (i.e., license plates and vehicle manufacturer) become an arduous task and beyond human capabilities. In this paper, we have developed a system using transfer learning-based DL techniques for automatic identification of Jordanian vehicles. The YOLOv3 (You Only Look Once) model was re-trained using transfer learning to accomplish the license plate detection, character recognition, and vehicle logo detection. While VGG16 (Visual Geometry Group) model was retrained to accomplish the vehicle logo recognition. To train and test these models, four datasets have been collected. The first dataset consists of 7,035 Jordanian vehicle images, the second dataset consist of 7,176 Jordanian license plates, and the third dataset consists of 8,271 Jordanian vehicle images. These datasets have been used to train and test the YOLOv3 model for Jordanian license plate detection, character recognition, and vehicle logo detection, respectively. While the fourth dataset consists of 158,230 vehicle logo images used to train and test the VGG16 model for the vehicle logo recognition. Text measures were used to evaluate the performance of our developed system. Moreover, mean average precision (mAP) measure was used to evaluate the YOLOv3 model of the detection tasks (i.e., license plate detection and vehicle logo detection). For license plate detection, the precision, recall, F-measure, and mAP were 99.6%, 100%, 99.8%, and 99.9%, respectively. While for character recognition, the precision, recall, and F-measure were 100%, 99.9%, and 99.95%, respectively. The performance of license plate recognition stage was evaluated by evaluating these two sub-stages as a sequence, where the precision, recall, and F-measure were 99.8%, 99.8%, and 99.8%, respectively. Furthermore, for vehicle logo detection, the precision, recall, F-measure, and mAP were 99%, 99.6%, 99.3%, and 99.1%, respectively, while for vehicle logo recognition, the precision, recall, F-measure were 98%, 98%, and 98%, respectively. The performance of vehicle logo recognition stage was evaluated by evaluating these two sub-stages as a sequence, where the precision, recall, and F-measure were 95.3%, 99.5%, and 97.4%, respectively.

Digital Twin Cities (DTCs) play a fundamental role in city planning and management. They allow three-dimensional modeling and simulation of cities. 3D semantic segmentation is the foundation for automatically creating enriched DTCs, as well as their updates. Past studies indicate that prior level fusion approaches demonstrate more promising precisions in 3D semantic segmentation compared to point level fusion, features level fusion, and decision level fusion families. In order to improve point cloud enriched semantic segmentation outcomes, this article proposes a new approach for 3D point cloud semantic segmentation through developing and benchmarking three prior level fusion scenarios. A reference approach based on point clouds and aerial images was proposed to compare it with the different developed scenarios. In each scenario, we inject a specific prior knowledge (geometric features,classified images ,etc) and aerial images as attributes of point clouds into the neural network’s learning pipeline. The objective is to find the one that integrates the most significant prior knowledge and enhances neural network knowledge more profoundly, which we have named the "smart fusion approach". The advanced Deep Learning algorithm "RandLaNet" was adopted to implement the different proposed scenarios and the reference approach, due to its excellent performance demonstrated in the literature. The introduction of some significant features associated with the label classes facilitated the learning process and improved the semantic segmentation results that can be achievable with the same neural network alone. Overall, our contribution provides a promising solution for addressing some challenges, in particular more accurate extraction of semantically rich objects from the urban fabric. An assessment of the semantic segmentation results obtained by the different scenarios is performed based on metrics computation and visual investigations. Finally,the smart fusion approach was derived based on the obtained qualitative and quantitative results.

The new energy structure needs to balance energy security and dual carbon goals, which has brought major challenges to coal-fired power plants, the pollution reduction and carbon emissions reduction of coal-fired power plants will be a key task in the future. This article proposes an optimization operation of electrostatic precipitators(ESP), based on the working mechanism and historical data of ESP, voltage-current characteristic model and outlet dust concentration prediction model are constructed by deep learning algorithm, Particle Swarm Optimization(PSO) is used to achieve the optimal energy consumption while ensuring stable outlet dust concentration. By training with historical data collected on site, accurate prediction of secondary current and outlet dust concentration of ESP has been achieved, the Mean Absolute Percentage Error(MAPE) of secondary current is less than 1.5%, and the MAPE of outlet dust concentration of ESP on the test set is less than 5.2%. Finally, the optimization experiment is carried out in a 330MW coal-fired power plants, the results showed that the fluctuation of the outlet dust concentration is more stable, and the energy saving is about 43% after optimization, according to the annual operation of 300 days, the annual average carbon reduction is approximately 2621.34 tons. This method is effective and can be applied widely.

Deep learning has been applied in many areas that have had a significant impact on applications that improves real-life challenges. The success of deep learning in a wide range of areas is due in part to the use of activation functions, which are particularly effective at solving non-linear problems. Activation functions are a key focus for researchers in artificial intelligence who aim to improve the performance of neural networks. This article provides a comprehensive explanation and comparison of different activation functions, with a focus on the arc tangent and its variations specifically. The paper presents experimental results that show that variations of the arc tangent using irrational numbers such as pi, the golden ratio, and Euler’s number, as well as a self-arctan function, produce promising results. Since we experimented with promising activation functions on two different problems, and datasets, we reached a result that different irrationals work well for different problems. In other words, arctan ϕ gives the best results mostly for multiclass classification and arctan e gives the best results for time series prediction problems. The paper focuses on a multi-class classification problem applied to the Reuters Newswire dataset and a time-series prediction problem on Türkiye energy trade value to show the impacts of activation functions.