Bearing fault diagnosis is crucial for ensuring the stable operation of mechanical equipment. With the continuous development of deep learning technology, Convolutional Neural Networks (CNNs) have demonstrated significant advantages in the field of fault diagnosis. This paper proposes a new method that combines various CNN architectures to improve the accuracy of bearing fault diagnosis. We designed five different convolutional network structures, including SerConv, ResConv, One-Shot Aggregation Convolution (OSAConv), Cross-Stage Aggregation Convolution (CSAConv), and MD-DAConv. Experimental results on the Case Western Reserve University (CWRU) bearing dataset show that the proposed method exhibits high accuracy and robustness in fault diagnosis. The results indicate that strategies such as multi-directional, multi-scale, and residual connections play a crucial role in enhancing the depth and breadth of feature extraction, while simple and effective feature fusion and information transmission mechanisms are key to ensuring the robustness and generalization ability of the model.

Identification of plants is a challenging task which aims to identify the family, genus, and species level according to morphological features. Automated deep learning-based computer vision algorithms are widely used for identifying plants and can help users to narrow down the possibilities. However, numerous morphological similarities between and within species make the classification difficult. In this paper, we tested a custom convolution neural network (CNN) and vision transformer (ViT) based models using the PyTorch framework to classify plants. We used a large dataset of 88K and 16K images for classifying plants at genus and species levels respectively. Our results show that for classifying plants at the genus level, ViT models perform better compared to CNN-based models ResNet50 and ResNet-RS-420, and other state-of-the-art CNN-based models suggested in previous studies on a similar dataset. The ViT model achieved top accuracy of 83.3% for classifying plants at the genus level. ViT models also perform better for classifying plants at the species level compared to CNN-based models ResNet50 and ResNet-RS-420, with a top accuracy of 92.5%. We show that the correct set of augmentation techniques plays an important role in classification success.

Alcoholism is attributed to regular or excessive drinking of alcohol and leads to the disturbance of the neuronal system in the human brain. This results in certain malfunctioning of neurons that can be detected by an electroencephalogram (EEG) using several electrodes on a human skull at appropriate positions. It is of great interest to be able to classify an EEG activity as that of a normal person or an alcoholic person using data from the minimum possible electrodes (or channels). Due to the complex nature of EEG signals, accurate classification of alcoholism using only a small data is a challenging task. Artificial neural networks, specifically convolutional neural networks (CNN), provide efficient and accurate results in various pattern-based classification problems. In this work, we apply CNN on raw EEG data, and demonstrate how we achieved 98% average accuracy by optimizing a baseline CNN model and outperforming its results in a range of performance evaluation metrics on the UCI-KDD EGG dataset. This article explains the step-wise improvement of the baseline model using the dropout, batch normalization, and kernel regularization techniques, and provides a comparison of the two models that can be beneficial for aspiring practitioners who aim to develop similar classification models in CNN. A performance comparison is also provided with other approaches using the same dataset.

Objectives: In order to solve the problem that most of the existing research focuses on the binary tactile attributes of objects,which ignores the tactile attribute strength and category recognition,an attribute strength and category recognition method based on convolutional neural network matrix-label is proposed. Methods:Firstly,in the data preparation stage,we preprocess the raw data and determine the matrix labels to build the haptic dataset.Secondly,in the feature extraction stage,we fuse the haptic data of two fingers and use the convolutional neural network to extract the attribute strength features.Finally,in the attribute strength and category recognition stage,all channel haptic data is fused to predict the attribute strength and category.Results:We compared with the multi-label convolutional neural network method in terms of elastic strength,hardness strength and category,and compared the attribute strength recognition capabilities of the two methods using novel objects outside the haptic dataset.The results show that the accuracy of the last 20 iterations of the matrix-label method has an average elastic strength of 96.73%,hardness strength of 97.34%,and category of 96.67%.The performance is better.When the Euclidean distance between the prediction of the novel object and the real label is less than 1,the accuracy of the elastic strength is best to reach 100%,and the hardness strength is best to reach 100%.The performance is better. Conclusions:The effectiveness of the method has been verified.Comparing with the convolutional neural network method,our method can effectively recognize the attribute strength and category of objects.

Cyber security plays an important role to protect our computer, network, program and data from unauthorized access. Intrusion detection system (IDS) and intrusion prevention system (IPS) are two main categories of cyber security, designed to identify any suspicious activities present in inbound and outbound network packets and restrict the suspicious incident. Deep neural network plays a significant role in the construction of IDS and IPS. This paper highlights a novel IDS using optimized convolution neural network (CNN-IDS). An optimized CNNIDS model is an improvement over CNN which selects the best weighted model by considering the loss in every epoch. All the experiments have been conducted on the well known NSL-KDD dataset. Information gain has been used for dimensionality reduction. The accuracy of the proposed model is evaluated through optimized CNN for both binary and multiclass categories. Finally, a critical comparison has been performed with other general classifiers like J48, Naive Bayes, NB tree, Random forest, Multilayer Perceptron (MLP), Support Vector Machine (SVM), Recurrent Neural Network (RNN) and Convolution Neural Network(CNN). All the experimental results demonstrate that the optimized CNN-IDS model records the best recognition rate with minimum model construction time.

This paper presents a novel approach for semantic segmentation of building roof in dense urban environment with Deep Convolution Neural Network (DCNN) using imagery acquired by a Chinese Very High Resolution (VHR) satellite mission, i.e. GaoFen-2 (GF-2). To provide an operational end-to-end work flow for accurate build roof mapping with feature extraction as well as image segmentation, a fully convolutional DCNN with both convolutional and deconvolutional layers is designed to perform the VHR image analysis for labeling pixels. Since the diverse urban patterns and building styles in large areas, sample image data sets of building roof and non-building roof are collected over different metropolitan regions in China. We selected typical cities with dense urban environment in each metropolitan region as study areas for collecting training and test samples. High performance cluster with GPU-mounted workstations is employed to perform the model training and optimization. With the building roof samples collected over different cities, the predictive model with multiple NN layers is developed for building roof labeling. The validation of the building roof map shows that the overall accuracy(OA) and the mean Intersection Over Union( mIOU) of DCNN based segmentation are 94.67%, 0.85 respectively, while CRF-refined segmentation achieved OA of 94.69% and mIOU of 0.83. The results suggest that the proposed approach is a promising solution for building roof mapping with VHR images over large areas across different urban and building patterns. With the operational acquisition of GF2 VHR imagery, it is expected to develop an automated pipeline for operational built-up area monitoring and timely update of building roof map over large areas.

Accurate and efficient traffic incident detection methods can effectively alleviate traffic congestion caused by traffic incidents, prevent secondary accidents and improve the safety of urban road traffic.Aiming at the problems that the traditional machine learning event detection method cannot fully extract the parameter characteristics of traffic flow and is not suitable formulti-dimensional and non-linear massive data, we propose a new traffic event detection method(CNN-XGBoost).This method combines the respective advantages of Convolution Neural Network(CNN) and Extreme Gradation Boosting (XGBoost). Firstly, we preprocessed the original freeway traffic incident detection data set by constructing initial variable set, data normalization, data balance processing and dimension reorganization. Secondly,we use CNN network to automatically extract the deep features of event detection data, and use XGBoost as a classifier to classify the extracted features for expressway traffic event detection.Finally, we use the data set of Hangzhou expressway microwave detector in China to carry out simulation experiments on CNN-XGBoost. The experimental results show that compared with XGBoost, CNN, Support Vector Machine (SVM) and Gradient Boosting Decision Tree (GBDT) and other methods, CNN-XGBoost method can effectively improve the accuracy of expressway traffic event detection and has better generalization ability.

For Convolutional Neural Network (CNN), Depthwise Separable CNN (DSCNN) is a preferred architecture for Application Specific Integrated Circuit (ASIC) implementation on edge devices. It can benefit from a multi-mode approximate multiplier proposed in this work. The proposed approximate multiplier uses two 4-bit multiplication operations to implement a 12-bit multiplication operation by reusing the same multiplier array. With this approximate multiplier, sequential multiplication operations are pipelined in a modified DSCNN to fully utilize the PE array in the convolutional layer. This Approximate (A-DSCNN) was implemented on TSMC 40-nm CMOS process with a supply voltage of 0.9 V. At the clock frequency of 200 MHz, the design achieves 4.78 GOPs/mW while occupying 1.24 mm x 1.24 mm silicon area. Compared to conventional DSCNN implemented in a similar process node, the chip area and power consumption were reduced by 53% and 25%, while the throughput was improved by 17%.

Agriculture is one of the indispensable fields for the survival of mankind. Potatoes also have a significant role in the field of agriculture. The quality and quantity of potatoes are significantly impacted by several diseases, such as early blight and late blight, and manual interpretation of these leaf diseases is time-consuming and inconvenient. Fortunately, leaf appearance is used to detect diseases in potato plants. Productivity will considerably rise if the infections are detected early. Different types of image processing methods and machine learning methods are used for early recognition of those diseases from leaf images so that the product losses are decreased significantly. For the incredible and marvelous performance of CNN, it is the most popular deep learning method that is used immensely for the recognition of leaf diseases from images. Several pretrained deep learning models, such as VGG16, ResNet50, InceptionV3, MobileNetV2, Xception, and a deep learning model developed using CNN, are employed for potato leaf disease classification and recognition on the same dataset. The transfer learning technique is applied to the pretrained model and the data augmentation technique is applied to the Proposed CNN model for potato disease classification from leaf samples. Compared to pretrained models, the proposed CNN model offers the lowest loss and highest accuracy for potato leaf disease detection while using fewer parameters and layers. It achieves the best performance with a test accuracy of 99.33% compared to other Pretrained models used in the diagnosis of potato leaf disease.

River bed materials serve multiple environmental functions as a habitat for aquatic invertebrates and fishes. At the same time, the particle size of the bed material reflects the tractive force of the flow regime in a flood and provides useful information for flood control. The traditional river bed particle size surveys, such as sieving, require time and labor to investigate river bed materials. The authors proposed a method to classify aerial images taken by unmanned aerial vehicle (UAV) using convolutional neural networks (CNN), our previous study showed that terrestrial riverbed material could be classified with high accuracy. In this study, we attempted to classify riverbed materials distributed in shallow waters where the bottom can be seen from UAVs. After training the CNN to classify the images with the same grain size as being in the same class even if the surface flow types taken overlapping the riverbed material were different, the total accuracy reached 90.3%. Moreover, the proposed method was applied to the wide-ranging area to determine the distribution of the particle size. In parallel, the microtopography was surveyed using Lidar-UAV, and the relationship between the microtopography and particle size distribution was discussed. In the steep section, coarse particles were distributed and formed a rapid. Fine particles were deposited on the upstream side of those rapids, where the slope had become gentler due to the damming. There was good agreement between the microtopographical trends and the grain size distribution.

Brain tumors are frequently classified with high accuracy using convolutional neural networks (CNNs) and better comprehend the spatial connections among pixels in complex pictures. Due to their tiny receptive fields, the majority of deep convolutional neural network (DCNN)-based techniques overfit and are unable to extract global context information from more significant regions. While dilated convolution retains data resolution at the output layer and increases the receptive field without adding computation, stacking several dilated convolutions has the drawback of producing a grid effect. To handle gridding artifacts and extract both coarse and fine features from the images, this research suggests using a dilated parallel deep convolutional neural network (PDCNN) architecture that preserves a wide receptive field. To reduce complexity, initially, input images are resized and then grayscale transformed. Data augmentation has since been used to expand the number of datasets. Dilated PDCNN makes use of the lower computational overhead and contributes to the reduction of gridding artifacts. By contrasting various dilation rates, the global path uses a low dilation rate (2,1,1), while the local path uses a high dilation rate (4,2,1) for decremental even numbers to tackle gridding artifacts and extract both coarse and fine features from the two parallel paths. Using three different types of MRI datasets, the suggested dilated PDCNN with the average ensemble method performs better. The accuracy provided by the Multiclass Kaggle dataset-III, Figshare dataset-II, and Binary tumor identification dataset-I is 98.35%, 98.13%, and 98.67%, respectively. In comparison to state-of-the-art techniques, the suggested structure improves results by extracting both fine and coarse features, making it efficient.

Neural networks were treated as black boxes for a long time. Previous works have unearthed what aspects of an image were important for convolutional layers at different positions in the network. This was done using deconvolutional networks. In this paper, we examine how well a convolutional neural network performs when those convolutional layers which are relatively unimportant for a particular image (i.e., the image does not produce one of the strongest activations) are skipped in the training, validating, and testing process.

：Non-Euclidean data, such as social networks, and citation relationships between documents, has node information and structural information. Graph Convolutional Network(GCN) can automatically learn node features and association information between nodes. The core ideology of the graph convolutional network is to aggregate node information by using edge information, thereby generating a new node feature. In the process of updating node features, there are two core influencing factors. One is the number of neighboring nodes of the central node, the other is the contribution of the neighboring nodes to the central node. Due to the previous GCN methods not simultaneously considering the numbers and different contributions of neighboring nodes to the central node, we design the Adaptive Attention Mechanism(AAM). To further enhance the representational capability of the model, we utilize Multi-Head Graph Convolution(MHGC). Finally, we adopt cross-entropy(CE) loss function to describe the difference between the predicted results of node categories and the ground truth (GT). Combined with backpropagation, this ultimately achieves accurate node classification. Based on AAM, MHGC and CE, we contrive the novel Graph Adaptive Attention Network (GAAN). Experiments show that the classification accuracy has achieved outstanding performances on Cora, Citeseer and Pubmed datasets.

In recent years, there has been a growing interest in the prediction of facial age due to its diverse applications in fields such as security, entertainment, and healthcare. The current scourge of underaged voting in Nigeria is a problem and this research delves into the realm of facial age prediction by employing four well-known convolutional neural network (CNN) architectures, namely VGG16, ResNet50, Mobile Net, and VGG19 to predict chronological ages from facial pictures of person. The objective is to achieve precise age estimation from facial images, utilizing two datasets: UTKFace and CASIA African Facial Datasets. The results of this investigation are noteworthy. Specifically, the VGG16 model which demonstrated remarkable performance, yielding a Mean Absolute Error (MAE) of 1.76 when applied to the UTK-Face dataset. Additionally, when utilizing Mobile-Net, an unprecedented MAE of 1.10 was achieved for the Casia Africa Face Dataset. Notably, this marks the first instance of employing the dataset for facial age detection with CNNs, and this approach outperformed previous works, yielding the lowest MAE among all the studies reviewed.

Singing melody extraction is the task that identifies the melody pitch contour of singing voice from polyphonic music. Most of the traditional melody extraction algorithms are based on calculating salient pitch candidates or separating the melody source from the mixture. Recently, classification-based approach based on deep learning has drawn much attentions. In this paper, we present a classification-based singing melody extraction model using deep convolutional neural networks. The proposed model consists of a singing pitch extractor (SPE) and a singing voice activity detector (SVAD). The SPE is trained to predict a high-resolution pitch label of singing voice from a short segment of spectrogram. This allows the model to predict highly continuous curves. The melody contour is smoothed further by post-processing the output of the melody extractor. The SVAD is trained to determine if a long segment of mel-spectrogram contains a singing voice. This often produces voice false alarm errors around the boundary of singing segments. We reduced them by exploiting the output of the SPE. Finally, we evaluate the proposed melody extraction model on several public datasets. The results show that the proposed model is comparable to state-of-the-art algorithms.

The colossal depths of the deep neural network sometimes suffer from ineffective backpropagation of the gradients through all its depths. Whereas, The strong performance of shallower multilayer neural structures prove their ability to increase the gradient signals in the early stages of training which easily gets backpropagated for global loss corrections. Shallow neural structures are always a good starting point for encouraging the sturdy feature characteristics of the input. In this research, a shallow, deep neural structure called PrimeNet is proposed. PrimeNet is aimed to dynamically identify and encourage the quality visual indicators from the input to be used by the subsequent deep network layers and increase the gradient signals in the lower stages of the training pipeline. In addition to this, the layerwise training is performed with the help of locally generated errors which means the gradient is not backpropagated to previous layers, and the hidden layer weights are updated during the forward pass, making this structure a backpropagation free variant. PrimeNet has obtained state-of-the-art results on various image datasets, attaining the dual objective of (1) compact dynamic deep neural structure, which (2) eliminates the problem of backwards-locking. The PrimeNet unit is proposed as an alternative to traditional convolution and dense blocks for faster and memory-efficient training, outperforming previously reported results aimed at adaptive methods for parallel and multilayer deep neural systems.

Image segmentation is the process of partitioning an image into multiple segments where the goal is to simplify the representation of the image and make the image more meaningful and easier to analyze. In particular, semantic segmentation is an approach of detecting the classes of objects, based on each pixel. In the past, most semantic segmentation models were for only one single style, such as urban street views, medical images, or even Manga. In this paper, we proposed a semantic segmentation model called Dual-Pyramid Wide Residual Network (DPWRN) to solve the segmentation for cross-style datasets, which is suitable for diverse segmentation applications. The DPWRN integrated the Pyramid of Kernel paralleled with Dilation (PKD) and Multi-Feature Fusion (MFF) to improve the accuracy of segmentation. To evaluate the generalization of the DPWRN and its superiority over most state-of-the-art models, three datasets with completely different styles are tested in the experiments. Finally, the DPWRN verified its generalization and also showed its superiority over most state-of-the-art models.

In recent years, different deep learning frameworks were introduced for hyperspectral image (HSI) classification. However, the proposed network models have a higher model complexity and do not provide high classification accuracy if few-shot learning is used. This paper pre-sents an HSI classification method that combines random patches network (RPNet) and re-cursive filtering (RF) to obtain informative deep features. The proposed method first convolves image bands with random patches to extract multi-level deep RPNet features. Thereafter, the RPNet feature set is subjected to dimension reduction through principal component analysis (PCA) and the extracted components are filtered using the RF procedure. Finally, HSI spectral features and the obtained RPNet-RF features are combined to classify the HSI using a support vector machine (SVM) classifier. In order to test the performance of the proposed RPNet-RF method, some experiments were performed on three widely known datasets using a few training samples for each class and classification results were compared with those obtained by other advanced HSI classification methods adopted for small training samples. The comparison showed that the RPNet-RF classification is characterized by higher values of such evaluation metrics as overall accuracy and Kappa coefficient (https://github.com/UchaevD/RPNet-RF).

Existing methods of hyperspectral anomaly detection still face several challenges: 1) Due to the limitations of self-supervision, avoiding the identity mapping of anomalies remains difficult; 2) The ineffective interaction between spatial and spectral features leads to the insufficient utilization of spatial information; 3) Current methods are not adaptable to the detection of multi-scale anomaly targets. To address the aforementioned challenges, we proposed a blind spot network based on multi-scale blind spot convolution for HAD. The multi-scale mask convolution module is employed to adapt to diverse scales of anomaly targets, while the dynamic fusion module is introduced to integrate the advantages of mask convolutions at different scales. The proposed approach includes a spatial-spectral joint module and a background feature attention mechanism to enhance the interaction between spatial-spectral features, with a specific emphasis on highlighting the significance of background features within the network. Furthermore, we propose a preprocessing technique that integrates pixel shuffle down-sampling (PD) operation with spatial spectral joint screening to address the issue of anomalous identity map-ping and facilitate finite-scale mask convolution to adapt to the detection of more scale targets. The proposed approach was assessed on four real hyperspectral datasets comprising anomaly targets of different scales. The experimental results demonstrate the effectiveness and superior performance of the proposed methodology compared with nine state-of-the art methods.

Research on image-inpainting tasks has mainly focused on enhancing performance by augmenting various stages and modules. However, this trend does not consider the increase in the number of model parameters and operational memory, which increases the burden on computational resources. To solve this problem, we propose a Parametric Efficient Image InPainting Network (PEIPNet) for efficient and effective image-inpainting. Unlike other state-of-the-art methods, the proposed model has a one-stage inpainting framework in which depthwise and pointwise convolutions are adopted to reduce the number of parameters and computational costs. To generate semantically appealing results, we selected three unique components: spatially-adaptive denormalization (SPADE), dense dilated convolution module (DDCM), and efficient self-attention (ESA). The SPADE was adopted to conditionally normalize activations according to the mask to distinguish between damaged and undamaged regions. The DDCM was employed at every scale to overcome the gradient-vanishing obstacle and gradually fill-in pixels by capturing global information along the feature maps. The ESA was utilized to obtain clues from unmasked areas by extracting long-range information. In terms of efficiency, our model has the lowest operational memory compared with other state-of-the-art methods. Both qualitative and quantitative experiments demonstrate the generalized inpainting of our method on three public datasets: Paris StreetView, CelebA, and Places2.

Underwater robotics equipped with visual detection systems have the function of detecting un-derwater artifacts, which is of great significance to deep-sea archaeology. Underwater artifacts are located in complex environments with poor imaging conditions, and the targets have chal-lenging problems such as breakage, stacking, and sediment burial, which causes the marine cul-tural target detection failures. To solve these problems, this paper proposes an underwater cul-tural target detection algorithm based on the deformable deep aggregation network model for underwater robotics exploration. To fully extract the target feature information of underwater targets in complex environments, this paper designs a multi-scale deep aggregation network with deformable convolutional layers. Besides, the BAM attention module is designed for fea-ture optimization, which enhances the potential feature information of the target while weak-ening the background interference information. Finally, the target prediction is achieved through feature fusion at different scales. The proposed algorithm has been extensively validat-ed and analyzed on the collected underwater artifact datasets, and the precision, recall, and mAP of the algorithm have reached 93.1%, 91.4%, and 92.8%, respectively. In addition, the present al-gorithm has been practically deployed on the underwater robotics. In the deep-sea tests, the ar-tifact detection frame rate reaches up to 19 fps, which satisfies the real-time target detection。

Deep learning methods have been widely used in PolSAR image classification. To learn the polarimetric information, many deep learning methods expect to learn high-level semantic features from original PolSAR data. However, only original data cannot learn multiple scattering features and complex structures for extremely heterogeneous terrain objects. In addition, deep learning methods always cause edge confusion due to the high-level features. To overcome these shortages, we propose a double-channel CNN network combined with an edge-preserving MRF model(DCCNN-MRF) for PolSAR image classification. Firstly, to combine complex matrix data and multiple scattering features together, a double-channel convolution network(DCCNN) is developed, which consists of a Wishart-based complex matrix and multi-feature subnetworks. The Wishart-based complex matrix network can learn the statistical characteristics and channel correlation well, and the multi-feature network can learn high-level semantic features well. Then, a unified network framework is designed to fuse two kinds of features to enhance advantageous features and reduce redundant ones. Finally, an edge preserving MRF model is designed to combine with the DCCNN network. In the MRF model, a sketch map-based edge energy function is designed by defining adaptive weighted neighborhood for edge pixels. Experiments are conducted on four real PolSAR data sets with different sensors and bands. Experimental results demonstrate the effectiveness of the proposed DCCNN-MRF method.

Multivariate Time Series Classification (MTSC) is one of most important tasks in time series analysis, aiding in activities such as human motion recognition and medical diagnostics. Existing methods for MTSC do not explicitly model temporal differences and generalize the idea of temporal difference into a efficient temporal module. Additionally, existing methods are not yet able to capture cross-variable relationships well during network training. As a result, they are unable to achieve convincing feature representation, leading to suboptimal classification accuracy. In this paper, we propose a novel MTSC model called Temporal Difference and Cross-variate Fusion Network (TDCFN), which integrates a two-stream differential LSTM network and a cross-variate feature extraction network to enhance feature representation. TDCFN achieves superior classification accuracy by capturing dynamic temporal evolution and inter-variable relationships. The experimental results show that TDCFN can achieve competitive performance with state-of-the-art multivariate time series classification approaches.

When extract building from high resolution remote sensing image with meter/sub-meter accuracy, the shade of trees and interference of roads are the main factors of reducing the extraction accuracy. Proposed a Bayesian Convolutional Neural Networks(BCNET) model base on standard fully convolutional networks(FCN) to solve these problems. First take building with no shade or artificial removal of shade as Sample-A, woodland as Sample-B, road as Sample-C. Set up 3 sample libraries. Learn these sample libraries respectively, get their own set of feature vector; Mixture Gauss model these feature vector set, evaluate the conditional probability density function of mixture of noise object and roofs; Improve the standard FCN from the 2 aspect:(1) Introduce atrous convolution. (2) Take conditional probability density function as the activation function of the last convolution. Carry out experiment using unmanned aerial vehicle(UVA) image, the results show that BCNET model can effectively eliminate the influence of trees and roads, the building extraction accuracy can reach 97%.

In remote sensing domain, it is crucial to annotate semantics, e.g., river, building, forest, etc, on the raster images. Deep Convolutional Encoder Decoder (DCED) network is the state-of-the-art semantic segmentation for remotely-sensed images. However, the accuracy is still limited, since the network is not designed for remotely sensed images and the training data in this domain is deficient. In this paper, we aim to propose a novel CNN for semantic segmentation particularly for remote sensing corpora with three main contributions. First, we propose to apply a recent CNN call ``Global Convolutional Network (GCN)'', since it can capture different resolutions by extracting multi-scale features from different stages of the network. Also, we further enhance the network by improving its backbone using larger numbers of layers, which is suitable for medium resolution remotely sensed images. Second, ``Channel Attention'' is presented into our network in order to select most discriminative filters (features). Third, ``Domain Specific Transfer Learning'' is introduced to alleviate the scarcity issue by utilizing other remotely sensed corpora with different resolutions as pre-trained data. The experiment was then conducted on two given data sets: ($i$) medium resolution data collected from Landsat-8 satellite and ($ii$) very high resolution data called ``ISPRS Vaihingen Challenge Data Set''. The results show that our networks outperformed DCED in terms of $F1$ for 17.48% and 2.49% on medium and very high resolution corpora, respectively.

Land cover segmentation, a fundamental task within the domain of remote sensing, boasts a broad spectrum of application potential. In this article, we focus on the land cover segmentation tasks and complete the following research work: Firstly, to address the issues of uneven distribution of foreground and background and significant differences in target scales in remote sensing images, we propose a decoder called MDCFD based on multi-dilation rate convolution fusion. The decoder utilizes dilated convolution to expand the receptive field, enhancing the model's ability to capture global features and thus improving the model's ability to distinguish between foreground and background. Meanwhile, we design a multi-dilation rate convolution fusion module (MDCFM), which fuses the outputs of different dilation rate convolution layers. Secondly, aiming at the problems of diverse scenes, significant differences between categories, and many background interferences in remote sensing images, we propose a hybrid attention module called LKSHAM based on large kernel convolution. This module combines spatial attention and channel attention mechanisms and combines the two attention modules in series. By introducing large kernel convolution, the model's ability to extract contextual information is improved. At the same time, we adopt a strategy of decomposing large kernel convolution into multiple depthwise convolutions to reduce computational complexity. The improved network models designed by this paper can achieve an improvement of over 1.1% in the mIoU metric on segmentation tasks on the Potsdam and Vaihingen datasets.

Real time camouflaged object detection so far using CAMO or COD10K datset has been able to achieve mAP50 less than 64%. We present a fourier transfom and Milkyway filament extraction based method extracting shape and edges of camouflaged object, achieving 88%+ mAP50 while continuing to deliver real time detection using YOLOv8s based models. This allows for search and rescue of wilelife and humans in smoke, fog, natural disasters possible by combining low compute Yolov8s model on headup display for rescue workers such as firefighters, first reponders etc.

Many studies in the current literature annotate patterns in stock prices and use computer vision models to learn and recognize these patterns from stock price-action chart images. Additionally, current literature also use Long Short-Term Memory Networks to predict prices from continuous dollar amount data. In this study, we combine the two techniques. We annotate the consolidation breakouts for a given stock price data, and we use continuous stock price data to predict consolidation breakouts. Unlike computer vision models that look at the image of a stock price action, we explore using the convolution operation on raw dollar values to predict consolidation breakouts under a supervised learning problem setting. Unlike LSTMs that predict stock prices given continuous stock data, we use the continuous stock data to classify a given price window as breakout or not. Finally, we do a regularization study to see the effect of L₁, L₂, and Elastic Net regularization. We hope that combining regression and classification shed more light on stock market prediction studies.

Table detection is a preliminary step in extracting reliable information from tables in scanned document images. We present CasTabDetectoRS, a novel end-to-end trainable table detection framework that operates on Cascade Mask R-CNN, including Recursive Feature Pyramid network and Switchable Atrous Convolution in the existing backbone architecture. By utilizing a comparatively lightweight backbone of ResNet-50, this paper demonstrates that superior results are attainable without relying on pre and post-processing methods, heavier backbone networks (ResNet-101, ResNeXt-152), and memory-intensive deformable convolutions. We evaluate the proposed approach on five different publicly available table detection datasets. Our CasTabDetectoRS outperforms the previous state-of-the-art results on four datasets (ICDAR-19, TableBank, UNLV, and Marmot) and accomplishes comparable results on ICDAR-17 POD. Upon comparing with previous state-of-the-art results, we obtain a significant relative error reduction of 56.36%, 20%, 4.5%, and 3.5% on the datasets of ICDAR-19, TableBank, UNLV, and Marmot, respectively. Furthermore, this paper sets a new benchmark by performing exhaustive cross-datasets evaluations to exhibit the generalization capabilities of the proposed method.

Single Image Super Resolution (SSIR) is a problem in computer vision where the goal is 1 to create high-resolution images from low-resolution ones. It has important applications in fields 2 such as medical imaging and security surveillance. While traditional methods such as interpolation 3 and reconstruction-based models have been used in the past, deep learning techniques have recently 4 gained attention due to their superior performance and computational efficiency. This article proposes 5 an Autoencoder based Deep Learning Model for SSIR, in particular, a light model that uses fewer 6 parameters without compromising performance. The down-sampling part of the Autoencoder 7 mainly uses 3 by 3 convolution and has no subsampling layers. The up-sampling part uses transpose 8 convolution and residual connections from the down sampling part. The model is trained using a 9 subset of the VILRC ImageNet database. The model is evaluated using quantitative metrics PSNR, 10 SSIM as well as qualitative measures such as perceptual quality. PSNR and SSIM figures as high as 11 76.06 and 0.93 are reported.

Saif et al. (J. Math. Comput. Sci. 21 (2020) 127-135) considered modified Laplace transform and developed some of their certain properties and relations. Motivated by this work, in this paper, we define modified Sumudu transform and investigate many properties and relations including modified Sumudu transforms of the power function, sine, cosine, hyperbolic sine, hyperbolic cosine, exponential function, and function derivatives. Moreover, we attain two shifting properties and a scale preserving theorem for the modified Sumudu transform. We give modified inverse Sumudu transform and investigate some relations and examples. Furthermore, we show that the modified Sumudu transform is the theoretical dual transform to the modified Laplace transform.

Development of rovers and development of infrastructure which enables them to probe other planets (such as Mars) have sparked a lot of interest recently specially with increasing public attention in Moon and Mars program by National Aeronautics and Space Administration. This is designed to be achieved by various means such as advanced spectroscopy and artificial intelligent techniques such as deep learning and transfer learning to enable the rover to not only map the surface of planet but to get a detailed information about its chemical makeup in layers beneath (deep learning) and in areas around point of observation (transfer learning). In this work, which is part of a proposal, later approach is explored. A systematic strategy is presented which make use of aforementioned techniques developed for metallic glass matrix composites as benchmark and helps develop algorithms for chemistry mapping of actual Martian surface on Perseverance Rover launching shortly.

With the rapid development of space programs in various countries, the number of satellites in space is increasing, resulting in an increasingly complex space environment. Therefore, improving space object identification technology has become highly important. We proposes a method of applying deep learning to intelligent detection of space object. We utilize 49 authentic 3D satellite models including 16 scenarios to generate a dataset comprising 17,942 images, which contains over 500 actual satellite photos. Additionally, we acquired a substantial amount of annotated data using a semi-automatic labeling method, which resulted in significant labor cost savings, and obtained a total of 39,000 labels. We validate the feasibility of the dataset using YOLOv3 and YOLOv7 models. What's more, we optimize the YOLOv7 model by integrating deformable convolution RepPoint into the YOLOv7 backbone to obtain the YOLOv7-R model. Through training with these two models, experimental results show that YOLOv3 achieves an accuracy of 0.927, YOLOv7 reaches an accuracy of 0.964, and YOLOv7-R achieves the highest accuracy at 0.983. This provides an effective solution for intelligent space object detection.

Convolution forms one of the most essential operations for the FPGA-based hardware accelerator. However, the existing designs often neglect the inherent architecture of FPGA, which puts forward an austere challenge on hardware resource requirements. Even though some previous works have proposed approximate multipliers or convolution acceleration algorithms to deal with this issue, the inevitable accuracy loss and resource occupation easily lead to performance degradation. Toward this, we first propose two kinds of resource-efficient optimized accurate multipliers based on LUTs or carry chains. Then targeting FPGA-based platforms, a generic multiply-accumulate structure is constructed by directly accumulating the partial products produced by our proposed optimized radix-4 Booth multipliers without intermediate multiplication and addition results. Experimental results demonstrate that our proposed multiplier achieves a maximum 51% look-up-table (LUT) reduction compared to the Vivado area optimized multiplier IP. Furthermore, the convolutional process unit using the proposed structure achieves a 36% LUT reduction compared to existing methods. As case studies, the proposed method is applied to DCT transformer and LeNet to achieves hardware resource saving without loss of accuracy.

3D object detection is essential for an accurate and reliable autonomous driving system. Currently, the methods used by the state-of-the-art two-stage detectors are not flexible enough and their fea-ture extraction capabilities are very limited to cope effectively with the disorder and irregularity of point clouds. In this paper, we combine the advantages of both PV-RCNN and PAConv (Position Adaptive Convolution) to create a completely new network, FANet, in order to overcome the ir-regularity and disorder of point clouds. The convolution in our network builds convolutional ker-nels from a basic weight matrix, whose combined coefficients are learned adaptively by LearnNet from relative points. This network allows for flexible modeling of complex spatial variations and geometric structures in the 3D point cloud, enabling better extraction of point cloud features and producing high-quality 3D proposal boxes. Compared to other methods, FANet is superior in terms of 3D object detection accuracy. Extensive experiments on the KITTI dataset have shown a signif-icant improvement in our approach.

Panchromatic (PAN) images contain abundant spatial information that is useful for earth observation, but always suffer from low-resolution due to the sensor limitation and large-scale view field. The current super-resolution (SR) methods based on traditional attention mechanism have shown remarkable advantages but remain imperfect to reconstruct the edge details of SR images. To address this problem, an improved super-resolution model which involves the self-attention augmented WGAN is designed to dig out the reference information among multiple features for detail enhancement. We use an encoder-decoder network followed by a fully convolutional network (FCN) as the backbone to extract multi-scale features and reconstruct the HR results. To exploit the relevance between multi-layer feature maps, we first integrate a convolutional block attention module (CBAM) into each skip-connection of the encoder-decoder subnet, generating weighted maps to enhance both channel-wise and spatial-wise feature representation automatically. Besides, considering that the HR results and LR inputs are highly similar in structure, yet cannot be fully reflected in traditional attention mechanism, we therefore design a self augmented attention (SAA) module, where the attention weights are produced dynamically via a similarity function between hidden features, this design allows the network to flexibly adjust the fraction relevance among multi-layer features and keep the long-range inter information, which is helpful to preserve details. In addition, the pixel-wise loss is combined with perceptual and gradient loss to achieve comprehensive supervision. Experiments on benchmark datasets demonstrate that the proposed method outperforms other SR methods in terms of both objective evaluation and visual effect.

The increasing use of Convolutional Neural Networks (CNN) has brought about a significant transformation in numerous fields, such as image categorization and identification. In the development of a CNN model for classifying images of sericea lespedeza (SL; Lespedeza cuneata) from weed images, four architectures were explored: CNN-Model Variant 1, CNN-Model Variant 2, VGG16, and ResNet50. The CNN-Model Variant 1 demonstrated 100 % validation accuracy, while Variant 2 achieved 90.78% validation accuracy. Pre-trained models, like VGG16 and ResNet50, were also analyzed. In contrast, ResNet50's steady learning pattern indicated potential for better generalization. A detailed evaluation of these models revealed that Variant 1 achieved a perfect score in precision, recall, and F1-score, indicating superior optimization and feature utilization. Variant 2 presented a balanced performance, with metrics between 86% and 93%. The VGG16 mirrored the behavior of Variant 2, both maintaining around 90% accuracy, but ResNet50's results revealed a conservative approach for class 0 predictions. Overall, Variant 1 stood out in performance, while both Variant 2 and VGG16 showed balanced results. The reliability of CNN model Variant 1 is highlighted by the significant accuracy percentages, which demonstrate its potential for practical implementation in agriculture. Smartphone application for the identification of SL in a field-based trial has shown promising results with an accuracy of 98%-99%. Using a CNN model with batch normalization has the potential to play a crucial role in redefining and optimizing the management of undesirable vegetation in the future.

Taekwondo has evolved from a traditional martial art into an official Olympic sport. This study introduces a novel action recognition model tailored for Taekwondo unit actions, utilizing joint motion data acquired via wearable inertial measurement unit (IMU) sensors. The utilization of IMU sensor-measured motion data facilitates the capture of the intricate and rapid movements characteristic of Taekwondo techniques. The model, underpinned by a conventional convolutional neural network (CNN)-based image classification framework, synthesizes action images to represent individual Taekwondo unit actions. These action images are generated by mapping joint motion profiles onto the RGB color space, thus encapsulating the motion dynamics of a single unit action within a solitary image. To further refine the representation of rapid movements within these images, a time-warping technique was applied, adjusting motion profiles in relation to the velocity of the action. The effectiveness of the proposed model was assessed using a dataset compiled from 40 Taekwondo experts, yielding remarkable outcomes: an accuracy of 0.998, precision of 0.983, recall of 0.982, and an F1-score of 0.982. These results underscore the time-warping technique's contribution to enhancing feature representation, as well as the proposed method's scalability and effectiveness in recognizing Taekwondo unit actions.

This paper introduces an innovative approach to lipreading, leveraging a web application designed to generate subtitles for videos where the speaker's mouth is visible and a comprehensive literature review that precedes the discussion, encompassing a thorough examination of various lipreading methods employed over the past decade. Our method employs a powerful deep learning model, featuring a 3D-convolution network and bidirectional LSTM, enabling accurate sentence-level predictions based solely on visual lip movements. With an impressive accuracy of 97%, our model is trained using pre-segmented lips regions, transformed into animated GIFs for effective pre-training. This work stands as a significant contribution to the evolving landscape of lipreading research, offering a practical and accurate solution for real-world applications.

This paper introduces a fusion convolutional architecture for efficient learning of spatio-temporal features in video action recognition. Unlike 2D CNNs, 3D CNNs can be applied directly on consecutive frames to extract spatio-temporal features. The aim of this work is to fuse the convolution layers from 2D and 3D CNNs to allow temporal encoding with fewer parameters than 3D CNNs. We adopt transfer learning from pre-trained 2D CNNs for spatial extraction, followed by temporal encoding, before connecting to 3D convolution layers at the top of the architecture. We construct our fusion architecture, semi-CNN, based on three popular models: VGG-16, ResNets and DenseNets, and compare the performance with their corresponding 3D models. Our empirical results evaluated on the action recognition dataset UCF-101 demonstrate that our fusion of 1D, 2D and 3D convolutions outperforms its 3D model of the same depth, with fewer parameters and reduces overfitting. Our semi-CNN architecture achieved an average of 16 – 30% boost in the top-1 accuracy when evaluated on an input video of 16 frames.

Multi-task Gaussian processes (MTGPs) are a powerful approach for modeling structured dependencies among multiple tasks. Researchers on MTGPs have contributed to enhance this approach in various ways. Current MTGP methods, however, cannot model nonlinear task correlations in a general way. In this paper we address this problem. We focus on spectral mixture (SM) based kernels and propose an enhancement of this type of kernels, called multi-task generalized convolution spectral mixture (MT-GCSM) kernel. The MT-GCSM kernel can model nonlinear task correlations and mixtures dependency, including time and phase delay, not only between different tasks but also within a task at the spectral mixture level. Each task in MT-GCSM has its own generalized convolution spectral mixture kernel (GCSM) with a different number of convolution structures and all spectral mixtures from different tasks are dependent. Furthermore, the proposed kernel uses inner and outer full cross convolution between base spectral mixtures, so that the base spectral mixtures in the tasks are not necessarily aligned. Extensive experiments on synthetic and real-life datasets illustrate the difference between MT-GCSM and other kernels as well as the practical effectiveness of MT-GCSM.

Aiming at the problems of inaccurate segmentation of long object and information loss of small object in real-time semantic segmentation algorithm, this paper proposes a lightweight multi-branch real-time semantic segmentation network based on BiseNetV2. The new auxiliary branch makes full use of spatial details and context information to cover the long object in the field of view. Meanwhile, in order to ensure the inference speed of the model, the asymmetric convolution is used in each stage of the auxiliary branch to design a structure with low computational complexity. In the multi-branch fusion stage, the alignment and fusion module is designed to provide guidance information for deep and shallow feature mapping, so as to make up for the problem of feature misalignment in the fusion of information at different scales, and thus reduce the loss of small target information. In order to further improve the model’s awareness of key information, a global context module is designed to capture the most important features in the input data. The proposed network uses NVIDIA GeForce RTX 3080 Laptop GPU experiment, on the road street view data set Cityscapes and CamVid average occurring simultaneously ratio reached 77.1% and 78.4% respectively, with running speed of 127 frames/s respectively and 112 frames/s. The experimental results show that the proposed algorithm can achieve real-time segmentation and improve the accuracy significantly, showing good semantic segmentation performance

The detection accuracy of bridge superficial defect using the deep neural network approach decreases significantly under light variation and weak texture condition. To address these issues, an enhanced intelligent detection method based on the YOLOv8 deep neural network is proposed in this study. Firstly, multi-branch coordinate attention (MBCA) is proposed to improve the accuracy of coordinate positioning by introducing a global perception module in coordinate attention mechanism. Furthermore, a deformable convolution based on MBCA is developed to improve the adaptability for complex feature shapes. Lastly, the deformable convolutional network-attention-YOLO (DAC-YOLO) detection algorithm is formed by replacing the deep C2F structure in the YOLOv8 architecture with deformable convolution. A supervised dataset consisting of 4794 bridge surface damage images is employed to verify the proposed method, and results show that it achieves improvements of 2.0% and 3.4% in mAP and R. Meanwhile, the model complexity decreases by 1.2G, increasing the detection speed by 3.5/f·s-1.

Vibration signal analysis is regarded as a fundamental approach in diagnosing faults in rolling bearings, and recent advancements have shown notable progress in this domain. However, the presence of substantial background noise often results in the masking of these fault signals, posing a significant challenge for researchers. In response, an Adaptive Denoising Autoencoder (ADAE) approach is proposed in this paper. The data representations are learned by the encoder through convolutional layers, while the data reconstruction is performed by the decoder using deconvolutional layers. Both the encoder and decoder incorporate adaptive shrinkage units to simulate denoising functions, effectively removing interfering information while preserving sensitive fault features. Additionally, dropout regularization is applied to sparsify the network and prevent overfitting, thereby enhancing the overall expressive power of the model. To further enhance ADAE's noise resistance, shortcut connections are added. Evaluation using publicly available datasets under scenarios with known and unknown noise demonstrates that ADAE effectively enhances the signal-to-noise ratio in strongly noisy backgrounds, facilitating accurate diagnosis of faults in rolling bearings.

In recent years, there has been significant growth in the ubiquity and popularity of three dimensional(3D) point clouds, with an increasing focus on the classification of 3D point clouds. To extract richer features from point clouds, many researchers have turned their attention to various point set regions and channels within irregular point clouds. However, this approach has limited capability in attending to crucial regions of interest in 3D point clouds and may overlook valuable information from neighboring features during feature aggregation. Therefore, this paper proposes a novel 3D point cloud classification method based on global attention and adaptive graph convolution. The method consists of two main branches: the first branch com-putes attention masks for each point, while the second branch employs adaptive graph convolu-tion to extract global features from the point set. It dynamically learns features based on point interactions, generating adaptive kernels to effectively and precisely capture diverse relationships among points from different semantic parts. Experimental results demonstrate that the proposed model achieves 93.8% in overall accuracy and 90.8% in average accuracy on the ModeNet40 dataset.

Cerebrovascular and airway structures are tubular structures used for transporting blood and gases, respectively, providing essential support for the normal activities of the human body. Accurately segmenting these tubular structures is the basis of morphology research and patho-logical detection. Nevertheless, accurately segmenting these structures from images presents great challenges due to their complex morphological and topological characteristics. To address this challenge, this paper proposes a framework UARAI based on the U-Net multi-scale reverse at-tention network and sparse convolution network. The framework utilizes a multi-scale structure to effectively extract the global and deep detail features of vessels and airways. Further, it enhances the extraction ability of fine-edge features by a joint reverse attention module. In addition, the sparse convolution structure is introduced to improve the features' expression ability without increasing the model's complexity. Finally, the proposed training sample cropping strategy re-duces the influence of block boundaries on the accuracy of tubular structure segmentation. The experimental findings demonstrate that the UARAI-based metrics, namely Dice and IoU, achieve impressive scores of 90.31% and 82.33% for cerebrovascular segmentation and 93.34% and 87.51% for airway segmentation, respectively. Compared to commonly employed segmentation tech-niques, the proposed method exhibits remarkable accuracy and robustness in delineating tubular structures such as cerebrovascular and airway. These results hold significant promise in facilitating medical image analysis and clinical diagnosis, offering invaluable support to healthcare profes-sionals.

Let be a power function $f_{r,M}(s)$ defined for every $s$ within the finite set $M$ as follows $$f_{r,M}(s)= \begin{cases} s^r, \ &s\in M,\\ 0, \ &\mathrm{otherwise}. \end{cases} $$ Let a discrete convolution of $f_{r,M}(s)$ be denoted as follows $\mathrm{Conv}_{r,M}[n]=(f_{r,M}*f_{r,M})[n]$. Let a real coefficients $A_{m,j}$ be given by the following recurrence $$ A_{m,j} = \begin{cases} 0, & \mathrm{if } \ j<0 \ \mathrm{or } \ j>m, \\ (2j+1)\binom{2j}{j} \sum_{d=2j+1}^{m} A_{m,d} \binom{d}{2j+1} \frac{(-1)^{d-1}}{d-j} B_{2d-2j}, & \mathrm{if } \ 0 \leq j < m, \\ (2j+1)\binom{2j}{j}, & \mathrm{if } \ j=m. \end{cases} $$ In this paper we show that for every $n>0$ the following odd-power identities involving coefficients $A_{m,j}$ and convolution transform $\mathrm{Conv}_{r,M}[n]$ hold $$ \begin{split} n^{2m+1}+1&=\sum_{r=0}^{m}A_{m,r}\mathrm{Conv}_{r,\mathbb{N}}[n],\\ n^{2m+1}-1&=\sum_{r=0}^{m}A_{m,r}\mathrm{Conv}_{r,\mathbb{Z}_{>0}}[n],\\ n^{2m+1}&=\sum_{r=0}^{m}A_{m,r}\sum_{k=1}^{n} k^r(n-k)^r\\ &=\sum_{r=0}^{m}A_{m,r}\sum_{k=0}^{n-1} k^r(n-k)^r. \end{split} $$

Late-onset Alzheimer’s disease (LOAD) is a dementia category disease that starts after the age of 65. Progressive problems in thinking, behaviour, learning etc. are its characteristics. Due to its progressive nature, early diagnosis is crucial. A complex pathway mechanism with different risk factors is involved in LOAD progression. A comprehensive study is done on factors causing LOAD, biomarkers used for diagnosis in clinical and preclinical stage of the disease, ML (Machine Learning) techniques to identify disease stage. We reviewed research papers on clinical and preclinical diagnosis of LOAD using ML techniques from biomarker data: MRI, PET image, MMSE, Genome / Gene expression, DNA / RNA binding sites with protein, Epigenetic expression /DNA, methylation, Speech and text. We have listed improvement scope after finding the gaps and scope of using Generative AI and SMOTE (Synthetic Minority Oversampling Technique) for data augmentation. At the end, we have elaborated current use of DL (Deep Learning) and CNN (‘Convolution Neural Network’) in Alzheimer’s disease (AD) diagnosis.

In high-traffic harbor waters, marine radar frequently encounters signal interference stemming from various obstructive elements, thereby presenting formidable obstacles in the precise identification of ships. To achieve precise pixel-level ship identification in the complex environments, a customized neural network-based ship segmentation algorithm named MrisNet is proposed. MrisNet employs a lightweight and efficient FasterYOLO network to extract features from radar images at different levels, capturing fine-grained edge information and deep semantic features of ship pixels. To address the limitation of deep features in the backbone network lacking detailed shape and structured information, an adaptive attention mechanism is introduced after the FasterYOLO network to enhance crucial ship features. To fully utilize the multi-dimensional feature outputs, MrisNet incorporates a Transformer structure to reconstruct the PANet feature fusion network, allowing for the fusion of contextual information and capturing more essential ship information and semantic correlations. In the prediction stage, MrisNet optimizes the target position loss using the EIoU function, enabling the algorithm to adapt to ship position deviations and size variations, thereby improving segmentation accuracy and convergence speed. Experimental results demonstrate that MrisNet achieves high recall and precision rates of 94.8% and 95.2%, respectively, in ship instance segmentation, outperforming various YOLO and other single-stage algorithms. Moreover, MrisNet has a model parameter size of 13.8M and real-time computational cost of 23.5G, demonstrating notable advantages in terms of convolutional efficiency. In conclusion, MrisNet accurately segments ships with different spot features and under diverse environmental conditions in marine radar images. It exhibits outstanding performance, particularly in extreme scenarios and challenging interference conditions, showcasing robustness and applicability.

Early detection of breast cancer, one of the most common cancers in women worldwide, raises survival rates and lowers the cost of treatment. Although breast cancer detection and classification (CAD) tools have improved, some issues and restrictions still require further research. The creation of breast cancer CAD systems, particularly with deep learning models, was significantly impacted by recent advancements in machine learning and image processing techniques. The current deep learning-based CAD system to identify and categorize masses in mammography is presented in a structured manner in this survey. The survey provides the most popular breast cancer CAD system evaluation metrics, publicly available mammographic datasets, and current image modalities. The survey examines the available literature, emphasizing its strengths and limitations. The survey also emphasizes the difficulties and drawbacks of the current methods for classifying and detecting breast cancer. We point out research gaps and make suggestions for further study. This systematic review may be beneficial for clinicians using CAD systems to diagnose breast cancer early on and for researchers looking for knowledge gaps and making more contributions to the field.

In recent years, intelligent fault diagnosis algorithms using deep learning method have achieved much success. However, the signals collected by sensors contain a lot of noise, which will have a great impact on the accuracy of the diagnostic model. To address this problem, we propose a one-dimensional convolutional neural network with multi-scale kernels (MSK-1DCNN) and apply this method to bearing fault diagnosis. We use a multi-scale convolution structure to extract different fault features in the original signal, and use the ELU activation function instead of the ReLU function in the multi-scale convolution structure to improve the anti-noise ability of MSK-1DCNN; then we use the training set with pepper noise to train the network to suppress overfitting. We use the Western Reserve University bearing data to verify the effectiveness of the algorithm and compare it with other fault diagnosis algorithms. Experimental results show that the improvements we proposed have effectively improved the diagnosis performers of MSK-1DCNN under strong noise and the diagnosis accuracy is higher than other comparison algorithms.

Since smoke usually occurs before a flame arises, fire smoke detection is especially significant for early warning systems. In this paper, a DSATA(Depthwise Separability And Target Awareness) algorithm based on depthwise separability and target awareness is proposed. Existing deep learning methods with convolutional neural networks pretrained by abundant and vast datasets are always used to realize generic object recognition tasks. In the area of smoke detection, collecting large quantities of smoke data is a challenging task for small sample smoke objects. The basis is that the objects of interest can be arbitrary object classes with arbitrary forms. Thus, deep feature maps acquired by target-aware pretrained networks are used in modelling these objects of arbitrary forms to distinguish them from unpredictable and complex environments. In this paper, this scheme is introduced to deal with smoke detection. The depthwise separable method with a fixed convolution kernel replacing the training iterations can improve the speed of the algorithm to meet the enhanced requirements of real-time fire spreading for detecting speed. The experimental results demonstrate that the proposed algorithm can detect early smoke, is superior to the state-of-the-art methods in accuracy and speed, and can also realize real-time smoke detection.

This paper presents a novel method to extract local features, which instead of calculating local extrema computes global maxima in a discretized scale-space representation. To avoid obtaining precise scales by interpolation and to achieve perfect rotation invariance, two essential techniques, increasing the width of kernels in pixel and utilizing disk-shaped convolution template are adopted in this method. Since the size of a convolution template is finite and finite templates can introduce computational error into convolution, we sufficiently discuss this problem and work out an upper bound of the computational error. The upper bound is utilized in the method to ensure that all features obtained are computed under a given tolerance. Besides, the technique of relative threshold to determine features is adopted to reinforce the robustness for the scene of changing illumination. Simulations show that this new method attains high performance of repeatability in various situations including scale change, rotation, blur, JPEG compression, illumination change and even viewpoint change.

Traffic sign detection plays a pivotal role in autonomous driving systems. The intricacy of the detection model necessitates high-performance hardware. Real-world traffic environments exhibit considerable variability and diversity, posing challenges for effective feature extraction by the model. Therefore, it is imperative to develop a detection model that is not only highly accurate but also lightweight. In this paper, we proposed YOLO-ADual, a novel lightweight model. Our method leverages the C3Dual and Adown lightweight modules as replacements for CPS and CBL modules in YOLOv5. The Adown module effectively mitigates feature loss during downsampling while reducing computational costs. Meanwhile, C3Dual optimizes the processing power for kernel feature extraction, enhancing computation efficiency while preserving network depth and feature extraction capability. Furthermore, the inclusion of the CBAM module enables the network to focus on salient information within the image, thus augmenting its feature representation capability. Our proposed algorithm achieves a mAP@0.5 of 70.1% while significantly reducing the number of parameters and computational requirements to 51.83% and 64.73% of the original model, respectively. Compared to various lightweight models, our approach demonstrates competitive performance in terms of both computational efficiency and accuracy.

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.

Breast cancer stands as a predominant health concern for women globally. As mammography is the primary screening tool for breast cancer detection, improving the detection of breast cancer at screening could save more lives. This mammography re-view paper provides a systematic review of computer-aided techniques during a specific time frame for the segmentation and classification of microcalcification, evaluating image processing, machine learning, and deep learning techniques. The systematic review is meticulously carried out, adhering closely to the preferred reporting items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines. This article focuses on mammographic breast cancer detection approaches based on automated systems, discussed chronologically from 1970 through 2023. This article encompasses the breadth of artificial intelligence-based methods from the most primitive to the most sophisticated models. Image processing and machine learning-based methods are compre-hensively reviewed. The evaluation of a deep learning architecture based on self-extracted features for classification tasks demonstrated outclass performance. Large-scale datasets required for a broader and in-depth analysis of novel methods for breast cancer detection are also discussed in this article. This research work is aligned with the United Nations' sustainability de-velopment goals.

The purpose of the present paper is to introduce and investigate new subclasses of analytic function class of bi univalent function defined in open unit disk connected with a linear q-convolution operator, which are associated with the quasi-subordination. We find coefficients estimate |h_2 |,|h_3 | for functions in these subclasses. Several known and new consequences of these results are also pointed out.

In the last few decades, glaucoma became the second biggest leading cause of irreversible vision loss. Because of its asymptotic growth, it is not properly diagnosed until the relatively late stage. To stop the severe damage by glaucoma it is needed to detect glaucoma in its early stages. Surprisingly diabetes also be the greatest cause of glaucoma. In the modern era, artificial intelligence makes great progress in the medical image processing field. Image analysis based on machine learning gives a huge success in diagnosis glaucoma without any misdiagnosis. The aim of this proposed paper is to create an automated process that can detect glaucoma and diabetic retinopathy. Here various Machine Learning models are used and results of these methods are presented.

A method of direction-of-arrival (DoA) estimation for FMCW radars is presented. In addition to MUSIC and ESPRIT, which are well-known high-resolution DoA estimation algorithms, deep learning has recently emerged as a very promising alternative. It is proposed in this paper to use a 3D convolutional neural network (CNN) for DoA estimation. The 3D-CNN extracts from the radar data cube spectrum features of the region of interest (RoI) centered on the potential positions of the targets, thereby capturing the spectrum phase shift information, which corresponds to DoA, along the antenna axis. Finally, the results of simulations and field experiments are provided to demonstrate the superior performance, as well as the limitations, of the proposed 3D-CNN.

In this paper, we propose a robust video understanding model for activity recognition by learning the actor’s pair-wise correlations and relational reasoning, exploiting spatial and temporal information. In order to measure the similarity between the pair appearances and construct an actor relations map, the Zero Mean Normalized Cross-Correlation (ZNCC) and the Zero Mean Sum of Absolute Differences(ZSAD) is proposed to allow the Graph Convolution Network (GCN) to learn how to distinguish group actions. We recommend that MNASNet be used as the backbone to retrieve features. Experiments show a 38.50% and 23.7% reduction in training time in the 2-stage training process along with a 1.52% improvement in accuracy against traditional methods.

Laser-targeted weeding methods further enhance the sustainable development of green agriculture, with one key technology being the improvement of weed localization accuracy. Here, we propose an improved YOLOv8 instance segmentation based on bidirectional feature fusion and deformable convolution (BFFDC-YOLOv8-seg) to address the challenges of insufficient weed localization accuracy in complex environments with resource-limited laser weeding devices. Initially, by training on extensive datasets of plant images, the most appropriate model scale and training weights are determined, facilitating the development of a lightweight network. Subsequently, the introduction of the Bidirectional Feature Pyramid Network (BiFPN) during feature fusion effectively prevents the omission of weeds. Lastly, the use of Dynamic Snake Convolution (DSConv) to replace some convolutional kernels enhances flexibility, benefiting the segmentation of weeds with elongated stems and irregular edges. Experimental results indicate that the BFFDC-YOLOv8-seg model achieves a 4.9% increase in precision, an 8.1% increase in recall rate, and a 2.8% increase in mAP50 value to 98.8% on a vegetable weed dataset compared to the original model. It also shows improved mAP50 over other typical segmentation models such as Mask R-CNN, YOLOv5-seg, and YOLOv7-seg by 10.8%, 13.4%, and 1.8%, respectively. Furthermore, the model achieves a detection speed of 24.8 FPS on the Jetson Orin nano standalone device, with a model size of 6.8MB that balances between size and accuracy. The model meets the requirements for real-time precise weed segmentation, suitable for complex vegetable field environments and resource-limited laser weeding devices.

In this paper, a relatively new hyper convolution spectra for characterization liquid, viscoelastic and solid samples is presented. Special attention is paid to the biophysical characterization of epithelial tissues and early detection of cancer of the skin, cervix, colon, and oral cavity. The method is based on the use of two types of white light, diffuse and polarized, from the same LED light sources. Based on the subtraction of the intensity values of these two lights and using the hyper convolution algorithm, the ratio of paired and unpaired electrons of tissue is obtained. This ratio correlates well with the state of the tissue: normal, irritating, precancerous, and cancerous. First, a proof-of-concept studies were performed, and later clinical studies on samples from one thousand two hundred fifteen people were performed. It showed that hyper Opto-Magnetic Imaging Spectroscopy (OMIS) can detect epithelial tissue cancers with an accuracy of 85-96%. Bearing in mind that the OMIS method also acquires the biophysical characteristics of the deeper skin layers (epidermis, basement membrane, dermis) by analysing convolution spectra of RGB channels of both diffuse and polarized white light, it is possible to apply this method in cosmetics (skin hydration, collagen, and elastin states, etc.). During OMIS applications, we noticed places where improvements could be made to increase method’s accuracy using advanced machine learning methodologies.

Binarization of degraded documents is an important preprocessing task for various document analysis such as OCR and historical document analysis. Existing studies have applied various convolutional neural network (CNN) models and generative models for document binarization, but they do not show generalized performance for noise that the model has not seen and it suffers from extracting elaborate text strokes. In this paper, to overcome these challenges, we utilize latent diffusion model (LDM), which is known for high-quality image generation model, for the first time in document binarization. By utilizing the iterative diffusion-denoising process in latent space, it shows high-quality cleaned binarized image generation and high generalized performance through using both data distribution and time step while training. Additionally, we apply gated U-Net to the backbone network to preserve text strokes using trainable gating value. Gated convolution can extract elaborate text stroke by allowing the model to focus on text region by combining gating value and feature. Furthermore, we maximize the effectiveness of the proposed model by training it with a combination of LDM loss and pixel-level loss, which is suitable for the model structure. Experiments on H-DIBCO and DIBCO benchmark datasets show that the proposed model outperforms existing methods.

Event-related potentials (ERPs) are estimated by averaging time-locked single trial electroencephalography (EEG) signals in response to specific events or stimuli. Classifying ERPs accurately is a challenge because (a) single trials have poor signal-to-noise-ratios (SNRs) and (b) it is difficult to collect large single trial ensembles to generate high SNR ERPs for classifier training and testing. The m-subsample averaging (m-SA) strategy which generates small-sample ERPs by repeated averaging of a small number of single trials drawn without replacement, has been proposed as a solution to the two problems. An ERP formed by averaging m single trials is referred to as an m-ERP where m is referred to as the averaging parameter. In this study, we conduct thorough analyses of m-SA and focus on issues not addressed in previous studies to better understand the beneficial properties of m-SA and to further support its application for ERP classification. Specifically, we (a) analyze the improvement in SNR as a function of m using the mean-root-mean-square SNR and visual analyses of m-ERP plots with confidence intervals, (b) analyze the improvement in interclass separation as a function of m, (c) determine how the SNR and interclass separation analyses can help to select the averaging parameter m, (d) determine the number of distinct m-ERPs that can be drawn from a single-trial ensemble, and (e) determine several probabilities related to the generation of distinct m-ERPs. Furthermore, an extensive set of experiments are designed to analyze the performance of support vector machine and convolution neural network classifiers employing m-SA with various combinations of the averaging parameters used for generating the training and test sets. The results confirm that ERPs can be classified accurately using small subsample averaging. Most importantly, it is concluded that m-SA can be deployed in practice to accurately classify ERPs in brain activity research and in clinical applications without having to collect a prohibitively large number of single trials.

Automatic target detection of remote sensing images (RSI) plays an important role in military reconnaissance, disaster monitoring, and target rescue. The core task of remote sensing target detection is to judge the target categories and complete precise location. However, the existing target detection algorithms have limited accuracy and weak generalization capability for remote sensing images with complex backgrounds. To achieve accurate detection of different categories targets in remote sensing images, this study presents a novel feature enhancement single shot multibox detector (FESSD) algorithm for remote sensing target detection. The FESSD introduces feature enhancement module and attention mechanism into the convolution neural networks (CNN) model, which can effectively enhance the feature extraction ability and nonlinear relationship between different convolution features. Specifically, the feature enhancement module is used to extract the multi-scale feature information, and enhance the model nonlinear learning ability; the self-learning attention mechanism (SAM) is used to expand the convolution kernel local receptive field, which makes the model extract more valuable features. In addition, the nonlinear relationship between different convolution features is enhanced using the feature pyramid attention mechanism (PAM). The advantage of FESSD over other state-of-the-art target detection methods is validated by experiments on the presented seven-class target detection dataset (SD-RSI) and the public DIOR dataset.

The fractional Hilbert transform, a generalization of the Hilbert transform, has been extensively studied in the literature because of its extensive use in optics, engineering, and signal processing. In the present work, we aimed to expand the fractional Hilbert transform to a space of generalized functions known as Boehmians. We introduce a new fractional convolution operator for the fractional Hilbert transform to prove a convolution theorem similar to the classical Hilbert transform and also to extend the fractional Hilbert transform to Boehmians. We also construct a suitable Boehmian space on which the fractional Hilbert transform exists. Further, we investigate convergence of the fractional Hilbert transform for the class of Boehmians and discuss the continuity of the extended fractional Hilbert transform.

The paper demonstrates the analysis of Corona Virus Disease based on a CNN probabilistic model. It involves a technique for classification and prediction by recognizing typical and diagnostically most important CT images features relating to Corona Virus. The main contributions of the research include predicting the probability of recurrences in no recurrence (first time detection) cases at applying our proposed Convolution neural network structure. The Study is validated on 2002 chest X-ray images with 60 confirmed positive covid19 cases and (650 bacterial – 412 viral -880 normal) x-ray images. The proposed CNN compared with traditional classifiers with proposed CHFS feature extraction model. The experimental study has done with real data demonstrates the feasibility and potential of the proposed approach for the said cause. The result of proposed CNN structure has been successfully done to achieve 98.20% accuracy of covid19 potential cases with comparable of traditional classifiers.

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.

The current neural networks for tomato leaf disease recognition have problems such as large model parameters, long training time, and low model accuracy. To solve these problems, a lightweight convolutional neural network LBFNet is proposed in this paper.First, a lightweight convolutional neural network LBFNet is established as the base model. Secondly, a three-channel attention mechanism module is introduced to learn the disease features in tomato leaf disease images and reduce the interference of redundant features. Finally, a cascade module is introduced to increase the depth of the model, solve the gradient descent problem, and reduce the loss caused by increasing the depth of the model. The quantized pruning technique is also used to further compress the model parameters and optimize the model performance. The results show that the LBFNet model achieves 99.06% accuracy on the LBFtomato dataset, with a training time of 996s and a single classification accuracy of over 94%. Further training using the saved weight file after quantized pruning makes the model accuracy reach 97.66%. Compared with the base model, the model accuracy was improved by 28%, and the model parameters were reduced by 96.7% compared with the traditional Resnet50. It was found that LBFNet can quickly and accurately identify tomato leaf diseases in complex environments, providing effective assistance to agricultural producers.

The rapid growth of smartphone financial services raises the need for secure mobile authentication. Continuous authentication is a user-friendly way to strengthen the security of smartphones by implicitly monitoring a user’s identity through sessions. Mobile continuous authentication can be viewed as an anomaly detection problem in which models discriminate between one genuine user and the rest of the imposters (anomalies). In practice, complete imposter profiles are hardly available due to the time and monetary cost, while leveraging genuine data alone yields poor generalized models due to the lack of knowledge about imposters. To address this challenge, we recast continuous mobile authentication as a few-shot anomaly detection problem, aiming to enhance the inference robustness of unseen imposters by using partial knowledge of available imposter profiles. Specifically, we propose a novel deep learning-based model, namely Local Feature Pooling based Temporal Convolution Network (LFP-TCN), which directly models raw sequential mobile data, aggregating global and local feature information. In addition, we introduce a random pattern mixing augmentation to generate class-unconstrained imposter data for the training. The augmented pool enables characterizing various imposter patterns from limited imposter data. Finally, we demonstrate practical continuous authentication using score-level fusion, which prevents long-term dependency or increased model complexity due to extended re-authentication time. Experiments on two public benchmark datasets show the effectiveness of our method and its state-of-the-art performance.

Neural network regularized learning has garnered significant attention in recent years. We give a systematic investigation on the performance of regularized regression associated zonal translation networks. We propose the concept of Marcinkiewicz-Zygmund inequality Setting (MZIS) for the scattered nodes collected from the unit sphere. We show that, under the MZIS, the corresponding convolutional zonal translation network has reproducing property. Based on these facts,we propose a kind of kernel regularized regression learning framework and provide upper bound estimate for the learning rate with the kernel approach. We also give proof for the density of the zonal translation network with spherical Fourier analysis.We provide the approximation error with a K-functional.

Drugs-drugs interactions(DDI) are entities composed of different chemical substructures (functional groups). In existing methods that predict drug–drug interactions based on the usage of substructures, each node is considered the center of a substructure, and adjacent nodes eventually become centers of similar substructures, resulting in redundancy. Furthermore, the significant differ- ences in structure and properties among compounds can lead to unrelated pairings, making it difficult to integrate information. This heterogeneity negatively affects the prediction results. To address these issues, we propose a drug–drug interaction prediction method based on substructure signature learning (DDI-SSL). This method extracts useful information from local subgraphs surrounding drugs and effectively utilizes substructures to assist in predicting drug side effects. Additionally, a deep clustering algorithm is used to aggregate similar substructures, allowing any individual subgraph to be reconstructed using this set of global signatures. Furthermore, we developed a layer-independent collaborative attention mechanism to model the mutual influence between drugs, generating signal strength scores for each class of drugs to mitigate noise caused by heterogeneity. Finally, we evaluated DDI-SSL on a comprehensive dataset and demonstrated improved performance in DDI prediction compared to state-of-the-art methods.

In response to reducing the energy cost of unmanned surface vehicles (USVs) while overcoming the low accuracy problem in surface target detection, a lightweight detection algorithm with multi-scale feature fusion is proposed. Based on the popular one-stage lightweight Yolov7-tiny target detection model, a lightweight extraction module is designed first by introducing the multiscale residual module to reduce the number of parameters and computational complexity while improving accuracy. The Mish and SiLU activation functions are used to enhance network feature extraction. Second, the path aggregation network employs coordinate convolution to strengthen spatial information perception. Finally, the dynamic head, which is based on the at-tention mechanism, improves the representation ability of object detection heads without any computational overhead. According to the experimental findings, the proposed model has 22.1% fewer parameters than the original model, 15% fewer GFLOPs, a 6.2% improvement in mAP@0.5, a 4.3% rise in mAP@0.5:0.95, and it satisfies the real-time criteria. According to the research, the suggested lightweight water surface detection approach includes a lighter model, a simpler computational architecture, more accuracy, and a wide range of generalizability. It performs bet-ter in a variety of difficult water surface circumstances.

This article focuses on obtaining the analytical solutions for parabolic Volterra integro- differential equations in d-dimensional with different types frictional memory kernel. Based on theories of Laplace transform, Fourier transform, the properties of Fox-H function and convolution theorem, analytical solutions of the equations in the infinite domain are derived under three frictional memory kernel functions respectively. The analytical solutions are expressed by infinite series, the generalized multi-parameter Mittag-Leffler function, Fox-H function and convolution form of Fourier transform. In addition, the graphical representations of the analytical solution under different parameters are given for one-dimensional parabolic Volterra integro-differential equation with power-law memory kernel. It can be seen that the solution curves subject to Gaussian decay at any given moment.

Accurate knowledge of flow rates is essential for hydraulic systems, as it allows for calculating hydraulic power when combined with pressure measurements. This data is helpful in applications such as predictive maintenance. However, most flow rate sensors in fluid power systems operate invasively, disrupting the flow and producing inaccurate results, particularly for transient flow conditions. A common approach is calculating the flow rate using the pressure difference along a pipeline based on the Hagen-Poiseuille law. However, this method is limited to laminar, steady, incompressible flow. This paper presents a novel soft sensor with an analytical model for transient, compressible pipe flow based on two pressure signals, thus no actual volumetric flow rate sensor is required. The model is derived by solving the fundamental fluid equations in the Laplace domain and converting the solution back to the time domain to obtain a physical representation. The relationship between the pressure difference signal and flow rate is derived using the four-pole theorem. The resulting analytical solution, including a convolution integral with a weighting function, shows high accuracy for transient and steady flows. This new equation enables the development of a soft sensor capable of non-invasive flow rate measurements for various pipe flow conditions.

Deep Neural Network (DNN) inference demands substantial computing power, resulting in significant energy consumption. A large number of negative output activations in convolution layers are rendered zero due to the invocation of the ReLU activation function. This results in a substantial number of unnecessary computations that consume significant amounts of energy. This paper presents ECHO: Energy-efficient Computation Harnessing Onilne Arithmetic - A MSDF-based accelerator for DNN inference, designed for computation pruning, utilizing an unconventional arithmetic paradigm known as online/ most-significant digit first (MSDF) arithmetic which performs computations in a digit-serial manner. The MSDF digital serial computation of online arithmetic enables overlapped computation of successive operations leading to substantial performance improvements. The online arithmetic, coupled with a negative output detection scheme, facilitates early and precise recognition of negative outputs. This, in turn, allows for the timely termination of unnecessary computations, resulting in a reduction of energy consumption. The implemented design has been realized on the Xilinx Virtex-7 VU3P FPGA and subjected to a comprehensive evaluation through a rigorous comparative analysis involving widely used performance metrics. Experimental results demonstrate promising power and throughput improvements compared to contemporary methods. In particular, the proposed design achieved an average improvement in power consumption of up to 81%, 82.9%, and 40.6% for VGG-16, ResNet-18, and ResNet-50 workloads compared to the conventional bit-serial design, respectively. Furthermore, significant average speedups of 2.39×, 2.6×, and 2.42 were observed when comparing the proposed design to conventional bit-serial designs for VGG-16, ResNet-18, and ResNet-50 models respectively.

The article proposes asymptotic methods for calculating Sommerfeld integrals, which enable us to calculate the integral using the expansion of a function into an infinite power series at the saddle point, where the role of a rapidly oscillating function under the integral can be fulfilled either by an exponential or by its product by the Hankel function. The proposed types of Sommerfeld integrals are generalized on the basis of integral representations of the Hertz radiator fields in the form of the inverse Hankel transform with the subsequent replacement of the Bessel function by the Hankel function. It is shown that the numerical values of the saddle point are complex. During integration, auxiliary or so-called standard integrals, which contain the main features of the integrand function, were used. As a demonstration of the accuracy of the technique, a previously known asymptotic formula for the Hankel functions was obtained in the form of an infinite series. The proposed method for calculating Sommerfeld integrals can be useful in solving the half-space Sommerfeld problem. An example is the expression in the form of an infinite series for the magnetic field of reflected waves, obtained directly through the Sommerfeld integral (SI).

India ranks among the top ten nations in the world for grape production. Fungal pathogens inflict damage to crop plants in turn making cultivators bear huge economical losses. With an output of 1.21 million tons (about 2% of 57.40 million tons produced globally). 1.2% of the nation’s total fruit cropland is covered by grapes. But due to fungal diseases the effect of the yield produced ranges from 5-80% depending on the severity of diseases which will affect the yield of grape vineyard. In precision agriculture, new sensing technologies and artificial intelligence could be used to automatically identify grapevine and disease pest symptoms. Traditional manual disease-monitoring methods are inefficient, labor-intensive, and ineffective. Timely effective and precise evaluation of grape diseases is admitted as a critical step in the field management. In this paper, we are explaining about different optical sensing methods applied for RGB, Multispectral and Thermal cameras. Section-wise we will be describing environmental set up for image-aquation, data-preprocessing, different modelling methods, evaluation matrix, result, and reviewer’s comment.

The utilization of Pig Emotion Recognition (PER) driven by Artificial Intelligence (AI) promises to mitigate labor costs and alleviate stress among domestic pigs, thereby minimizing the need for consistent human intervention. Nevertheless, this research acknowledges the inherent limitations within the raw PER datasets, which often include irrelevant porcine features, hence impeding genuine progress in real-world evaluations. A significant proportion of PER datasets derive from sequential pig imagery obtained from video recordings, and a common pitfall in these studies is the unregulated shuffling of data. This lack of control can result in the overlap of data samples between training and testing groups, thereby yielding skewed experimental evaluations. To address these challenges, this paper introduces a novel solution in the form of the Semi-Shuffle-Pig Detector (SSPD) for PER datasets, with the intent to facilitate a less biased experimental output. By applying the SSPD, we can ensure that all testing data samples remain distinct from the training datasets, and any superfluous information from raw images is systematically discarded. This optimized method enhances the true performance of classification, providing unbiased experimental evaluations. Notably, our approach has led to a remarkable improvement in the Isolation After Feeding (IAF) metric by 20.2\% and achieved higher accuracy in segregating IAF and Paired After Feeding (PAF) classifications exceeding 92\%. This methodology, thereby, ensures the preservation of pertinent data within the PER system and eliminates potential biases in experimental evaluations. Consequently, it elevates the accuracy and reliability of real-world PER applications, resulting in a tangible positive impact on both pig welfare management and food safety standards.

You Only Look One (YOLO) and its variants have emerged as the most popular real-time object detection algorithms. They have been widely used in real-time smart transportation applications due to their low-latency detection and high accuracy. However, because of the diverse characteristics of YOLO models, choosing selecting the optimal model according to various applications and environments in smart transportation is critical. In this article, we conduct an empirical evaluation and analysis study for most YOLO versions to assess their performance in smart transportation. To achieve this, we first measure the average precision of YOLO models across multiple datasets (e.g., COCO, PASCAL VOC). Second, we analyze the performance of YOLO models on multiple object categories within each dataset, focusing on classes relevant to road transportation such as those commonly used in smart transportation applications. Third, multiple intersection-over-union (IoU) thresholds are considered in our performance measurement and analysis. By examining the performance of various YOLO models across datasets, IoU thresholds, and object classes, we make six observations on these three aspects while aiming to identify optimal models for road transportation scenarios. Thus, YOLOv5 and YOLOv8 outperform other models in all three aspects due to their novel performance features. For instance, YOLOv5 achieves stable performance thanks to its cross-stage partial darknet-53 (CSPDarknet53) backbone, auto-anchor mechanism, and efficient loss functions including IoU loss, complete IoU loss, focal loss, gradient harmonizing mechanism loss. Similarly, YOLOv8 outperforms others with its upgraded CSPDarknet53 backbone, anchor-free mechanism, and efficient loss functions like complete IoU loss and distribution focal loss.

Underwater natural gas pipelines constitute critical infrastructure for energy transportation. Any damage or leakage in these pipelines poses serious security risks, directly threatening marine and lake ecosystems and potentially causing operational issues and economic losses in the energy supply chain. Due to the difficulty in detecting deterioration over time and regularly inspecting these submerged pipelines by divers, the use of unmanned underwater vehicles (UUVs) becomes crucial in this field. In this study, an underwater pipeline tracking experiment was carried out by providing autonomous features to a remote-controlled unmanned underwater vehicle. During the tracking of the underwater pipeline, damages were identified, and the locations of these damages were determined. The navigation information of the underwater vehicle, including orientation in the x, y, z axes (roll, pitch, yaw) from a gyroscope integrated with magnetic compass, speed and position information in the three axes from an accelerometer, and the distance to the water surface from a pressure sensor, was integrated into the vehicle. Pre-tests determined the necessary pulse width modulation values for the thrusters of the vehicle, enabling autonomous operation by providing these values as input to the thruster motors. In this study, the vehicle moves in 3D was achived by the vertical thruster of the vehicle was activated to maintain a specific depth, and applying equal force to the vehicle’s right and left thrusters forward movement, while differential force induced deviation angles. In pool experiments, the unmanned underwater vehicle autonomously tracked the pipeline as intended, identifying damages on the pipeline using images captured by the vehicle's camera. The images for damage assessment were processed using a convolution neural network (CNN) algorithm, which is a deep learning method. The position of the damage relative to the vehicle was estimated from the pixel dimensions of the identified damage. The location of the damage relative to its starting point was obtained by combining these two positional pieces of information from the vehicle's navigation system. The all study was performed within the Python environment.

This paper proposed emotion recognition methods for consecutive facial images, and implements the inference of neural network model on a field-programmable gate array (FPGA). The proposed emotion recognition methods are based on a neural network model architecture that combines convolution neural networks (CNNs), long short-term memory (LSTM), and fully connected neural networks (FCNNs), called CLDNN or ConvLSTM-FCN. This type of neural network mod-el can analyze the local feature sequences obtained through convolution of data, making it suita-ble for processing time-series data such as consecutive facial images. In this paper, sequences of facial images are sampled from videos corresponding to various emotional state of subjects. The sampled images are then pre-processed with the processes includes facial detection, greyscale conversion, resize, and data augmentation if necessary. The 2-D CNN in ConvLSTM-FCN is used for feature extraction for these pre-processed facial images. These sequences of facial im-age features are time sequences with dependent properties between the elements within them. The LSTM is then used to model these time sequences followed by fully connected neural net-works (FCNNs) for classification. The proposed consecutive facial emotion recognition method achieves an average recognition rate of 99.51% on RAVDESS dataset, 87.80% on BAUM-1s dataset and 96.82% on eNTERFACE’05 data set, using 10-fold cross-validation on the PC. Some com-parisons of recognition accuracies between the proposed method and the other existing related works are conducted in this paper. According to the comparisons, the proposed emotion recog-nition methods outperform the existing related researches. The proposed emotion recognition methods are then implemented on an FPGA chip using the neural network model inference algo-rithms in this paper, and the accuracies of the experiments conducted on the FPGA chip are iden-tical to those obtained on the PC. This verifies that the proposed neural network model imple-mented on the FPGA chip performs well.

- Early and accurate detection of pneumonia from chest X-ray images is crucial for timely treatment and patient care. In study presents a robust computational framework leveraging the CheXnet algorithm, a deep learning model, to identify pneumonia from chest X-rays. Developed a neural network architecture trained on a substantial dataset of chest X- ray images labeled as 'Normal' and 'Pneumonia'. The user- friendly interface of the system allows for seamless uploading of images, with the model subsequently providing predictions, a confusion matrix, accuracy metrics, and layer activations visualizations. The algorithm's performance was rigorously evaluated, with a confusion matrix indicating a high number of true positives and true negatives. Specifically, the model detected pneumonia with an accuracy of 92.47%, as shown by the results displayed in the system's interface. The precision, recall, and F1-score for pneumonia detection were 0.90, 0.96, and 0.93 respectively, with the overall system achieving an accuracy of 0.91 across the test set of 624 images. Visualizations of the neural network layers reveal the complex feature mappings the model uses to differentiate between normal and pathological findings. These insights could be vital for enhancing the interpretability of deep learning models in medical diagnostics. The Results indicate that the CheXnet-based neural network is a promising tool for automating pneumonia detection in clinical settings, which can assist radiologists in the diagnostic process and potentially improve patient outcomes through earlier intervention.

This work is part of an ongoing research project to develop an unmanned flying social robot to monitor dependants at home in order to detect the person’s state and bring the necessary assistance. In this sense, this paper focuses on the description of a virtual reality (VR) simulation platform for the monitoring process of an avatar in a virtual home by a rotatory-wing autonomous unmanned aerial vehicle (UAV). This platform is based on a distributed architecture composed of three modules communicated through the Message Queue Telemetry Transport (MQTT) protocol: the UAV Simulator implemented in MATLAB/Simulink, the VR Visualiser developed in Unity, and the new emotion recognition (ER) System developed in Python. Using a face detection algorithm and a convolutional neural network (CNN), the ER System is able to detect the person’s face in the image captured by the UAV’s on-board camera and classify the emotion among seven possible ones (surprise, fear, happiness, sadness, disgust, anger or neutral expression). The experimental results demonstrate the correct integration of this new computer vision module within the VR platform, as well as the good performance of the designed CNN, with around 85% in the F1-score, a mean of the precision and recall of the model. The developed emotion detection system can be used in the future implementation of the assistance UAV that monitors dependent people in a real environment, since the methodology used is valid for images of real people.

In the paper, by virtue of convolution theorem for the Laplace transforms, Bernstein's theorem for completely monotonic functions, and other techniques, the author finds necessary and sufficient conditions for a difference constituted by four derivatives of a function involving trigamma function to be completely monotonic.

Tables in the document image are one of the most important entities since they contain crucial information. Therefore, accurate table detection can significantly improve information extraction from tables. In this work, we present a novel end-to-end trainable pipeline, HybridTabNet, for table detection in scanned document images. Our two-stage table detector uses the ResNeXt-101 backbone for feature extraction and Hybrid Task Cascade (HTC) to localize the tables in scanned document images. Moreover, we replace conventional convolutions with deformable convolutions in the backbone network. This enables our network to detect tables of arbitrary layouts precisely. We evaluate our approach comprehensively on ICDAR-13, ICDAR-17 POD, ICDAR-19, TableBank, Marmot, and UNLV. Apart from the ICDAR-17 POD dataset, our proposed HybridTabNet outperforms earlier state-of-the-art results without depending on pre and post-processing steps. Furthermore, to investigate how the proposed method generalizes unseen data, we conduct an exhaustive leave-one-out-evaluation. In comparison to prior state-of-the-art results, our method reduces the relative error by 27.57% on ICDAR-2019-TrackA-Modern, 42.64% on TableBank (Latex), 41.33% on TableBank (Word), 55.73% on TableBank (Latex + Word), 10% on Marmot, and 9.67% on UNLV dataset. The achieved results reflect the superior performance of the proposed method.

Change Detection (CD) of aerial images refers to identifying and analyzing changes between two or more aerial images of the same location taken at different times. The CD is a highly challenging task due to the need to distinguish relevant changes, such as urban expansion, deforestation, or post-disaster damage assessment, from irrelevant ones, such as light conditions, shadows, and seasonal variations. Many CD papers have recently been published, where most of the papers that proposed a new model contained a comparison between their proposed and state-of-the-art (SOTA) models. While many recent studies propose new deep learning (DL) models for improving CD performance, their comparative analyses are often restricted, lacking comprehensive insights into the proposed models' real-world generalizability, robustness, and performance trade-offs across diverse change characteristics. This paper presents a novel generic framework to systematically benchmark and assess DL-based CD models through three parallel pipelines: 1) cross-testing models on diverse benchmark datasets to evaluate generalization, 2) robustness analysis against different image corruptions, and 3) multi-faceted contour-level analytics evaluating detection sensitivity to change size/complexity. The framework is applied to comparatively evaluate five state-of-the-art DL-based CD models - Changeformer, BIT, Tiny, SNUNet, and CSA-CDGAN. Extensive experiments unveil each model's strengths, limitations and biases, highlighting their relative proficiencies in generalizing across data distributions, resilience to noise corruption, and discriminative capabilities for changes of varying characteristics. The proposed benchmarking framework demonstrates significant potential for guiding the selection of suitable CD models tailored to specific application requirements by comprehensively evaluating their generalizability, robustness, and detection capabilities across diverse real-world scenarios. This systematic evaluation approach can drive future research into developing more robust and versatile CD solutions aligned with practical needs.

Nowadays, power sector is an area that gather great scientific interest, due to events such as the increase in electricity prices in the wholesale energy market and new investments due to technological development in various sectors. These new challenges have in turn created new needs, such as the accurate prediction of the electrical load of the end users. On the occasion of the new challenges, Artificial Neural Networks approaches have become increasingly popular due to their ability to adopt efficiently to time-series predictions. In this paper, it is presented the development of a model which, through an automated process, will provide an accurate prediction of electrical load for the island of Thira in Greece. Through an automated application, deep learning load forecasting models have been created, such as Multilayer Perceptron, Long Short-Term Memory (LSTM), Convolutional Neural Network One Dimensional (CNN-1D), CNN-LSTM, Temporal Convolutional Network (TCN) and a proposed hybrid model called Convolutional LSTM Encoder-Decoder. The results in terms of prediction accuracy show satisfactory performances for all models, with the proposed hybrid model achieving the best accuracy.

In the paper, by convolution theorem for the Laplace transforms and analytic techniques, the author finds necessary and sufficient conditions for complete monotonicity, monotonicity, and inequalities of several functions involving polygamma functions. By these results, the author derives a lower bound of a function related to the sectional curvature of the manifold of the beta distributions. Finally, the author poses several guesses and open problems related to monotonicity, complete monotonicity, and inequalities of several functions involving polygamma functions.