During two decades back, the growth of the fisheries sector in Indonesia was showing increasing trend, however, behind the rapid development of the sector, small-scale fisheries role is often neglected as one of the main actors to support the whole sector. They remain poor and doing fishing activities traditionally. Therefore, this study aims to describe the real situation of small-scale fishers, analyzing fishers’ ownership of the boat, analyzing their income level to reveal their poverty condition, to analyze the factors affecting their income. The regression analysis results indicate that fishing days, boat category, sea fish catch, fish selling price, variable cost, and secondary job have significant affect to fishers’ net income.

Crack inspection in railway sleepers is crucial for ensuring rail safety and avoiding deadly accidents. Traditional methods for detecting cracks on railway sleepers are very time-consuming and lack efficiency. Therefore nowadays, researchers are paying attention to the vision-based algorithm, especially Deep Learning algorithms. In this work, we adopted the U-net for the first time for detecting cracks on a railway sleeper and proposed a modified U-net architecture named Dense U-net for segmenting the cracks. In the Dense U-net structure, we established several short connections between the encoder and decoder blocks, which enabled the architecture to obtain better pixel information flow. Thus, the model extracted the necessary information in more detail to predict the cracks. We collected images from railway sleepers, processed them in a dataset, and finally trained the model with the images. The model achieved an overall F1-score, precision, Recall, and IoU of 86.5% 88.53%, 84.63%, and 76.31% respectively. We compared our suggested model with the original U-net, and the results demonstrate that our model outperformed the U-net in both quantitative and qualitative results. Moreover, we considered the necessity of crack severity analysis and measured a few parameters of the cracks (e.g., length, maximum width, area, density). The engineers must know the severity of the cracks to have an idea about the most severe locations and take the necessary steps to repair the badly affected sleepers.

The implementation of residential distributed generation (DG) in South America has shown significant growth in the last decade, driven by increased research on photovoltaics and renewable energy sources. Regulatory policies and business models have been crucial in fostering the adoption of DG in the region, with residential self-supply being the most receptive sector for implementing the concept. This article presents a literature review on DG as applied to residential self-supply in South America by comprehensively analyzing documents published between 2013 and 2023, applying the PRISMA methodology. In total, 37 literature resources have been systematically selected and reviewed to contribute to the identification of trends and developments in the field of DG in the residential sector, as well as highlighting areas that require further research and attention from stakeholders, including policymakers, investors, technology providers and consumers in the South American context.

AK is a common precancerous skin condition that requires effective detection and treatment monitoring. To improve the monitoring of AK burden in clinical settings with enhanced automation and precision, the present study evaluates the application of semantic segmentation based on the U-Net architecture (AKU-Net). AKU-Net employs transfer learning to compensate for the relatively small dataset of annotated images and integrates a recurrent process based on convLSTM to exploit contextual information and address the challenges related to low contrast and ambiguous boundaries of AK-affected skin regions. We used an annotated dataset of 569 clinical photographs from 115 patients with actinic keratosis to train and evaluate the model. From each photograph, patches of 512x512 pixels were extracted using translation lesion boxes that encompass lesions in different positions and capture different contexts of perilesional skin. In total, 16488 translation augmented crops were used for training the model, and 403 lesion center crops were used for testing. To demonstrate the improvements in AK detection, AKU-net was compared with plain U-Net and U-Net++ architectures. The experimental results highlighted the effectiveness of AKU-Net, improving both automation and precision over existing approaches, paving the way for more effective and reliable evaluation of actinic keratosis in clinical settings.

Several strategies promote phyllosphere colonization by soil-born Trichoderma plant-beneficial strains. One of these is foliar spraying with suspensions containing large amounts of chlamydospores – spores with a thick cell wall structure that makes them highly resistant to harsh environmental conditions. Trichoderma biomass was produced by cultivation on cornmeal medium and compared by microscopic and thermogravimetric analysis with the biomass produced in potato dextrose broth. Both analyses revealed higher chlamydospores content and thicker mycelium in the fungal biomass produced on cornmeal medium. The Trichoderma suspension rich in chlamydospores was sprayed on bitter gourd (Momordica charantia) leaves at two concentration levels, 106 and 108 ufc/ml. The effect of these treatments on plant physiological parameters, leaf photosynthetic pigments, polyphenol and flavonoid contents, antioxidant activity of leaves and fruits and yield was compared to the control (plants sprayed with water) and to the experimental treatment sprayed with 108 ufc propagules produced in potato dextrose broth. The effect of chlamydospore-rich suspensions on plant physiological parameters was more pronounced and long-lasting compared with the other treatments. The treatment with chlamydospore-rich suspension enhanced the accumulation of polyphenols and flavonoids in the leaves and fruits and increased the antioxidant activity. The application of the foliar treatment with Trichoderma suspensions did not modify the cytocompatibility of the extracts from fruits determined on L929 cells.

Abstract: This study observes the characteristics and influencing factors of carbon fluxes of the Momoge salt marsh ecosystem over four years, which behaves as a CO2 sink. The daily, seasonal, and interannual variations of CO2 fluxes in the Momoge salt marshes were observed using the eddy covariance method and compared with various environmental factors. An overall daily “U” shaped distribution was observed, with uptake during the day (negative values) and release at night (positive values). Annually, the carbon fluxes in the study area roughly exhibit a “V” shape. Carbon fluxes during the non-growing season predominantly showed positive values, indicating the release of CO2 into the atmosphere. Photosynthetically active radiation was the primary influencing factor affecting hourly and daytime variations in net ecosystem exchange (NEE) during the growing season, while temperature was the main factor influencing nighttime NEE dynamics. Air temperature, soil temperature, photosynthetically active radiation, precipitation, and water level all had significant impacts on daily net CO2 exchange. At the monthly scale, larger values of soil temperature, air temperature, photosynthetically active radiation, and aboveground biomass correspond to a stronger carbon absorption capacity of the ecosystem. Overall, temperature remains the primary factor for carbon fluxes in the Momoge wetlands.

This research introduces a pioneering solution to the challenges posed by gastrointestinal tract (GI) cancer in radiation therapy, focusing on the imperative task of precise organ segmentation for minimizing radiation-induced damage. GI imaging has historically used manual demarcation, which is laborious and uncomfortable for patients. We address this by introducing the ResECA-U-Net deep learning model, a novel combination of the UNet and ResNet34 architectures. Furthermore, we further augment its functionality by incorporating the Efficient Channel Attention (ECA-Net) methodology. By utilizing data from the UW-Madison Carbone Cancer Centre, we carefully investigate several image processing techniques designed to capture critical local characteristics. With its foundation in computer vision concepts, the ResECA-U-Net model is excellent at extracting fine details from GI images. Sophisticated metrics such as intersection over union (IoU) and the dice coefficient are used to evaluate performance. Our study's outcomes demonstrate the effectiveness of the suggested method, yielding an impressive 96.27% Dice coefficient and 91.48% IoU. These results highlight the significant contribution that our strategy has made to the advancement of cancer therapy. Beyond its scientific merits, this work has the potential to significantly enhance cancer patients' quality of life and provide better long-term outcomes. Our work is a significant step towards automating and optimizing the segmentation process, which can potentially change how GI cancer is treated completely.

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

Public health suffers from noxious gas of beached Sargassum algae. Floating barrages containing them at sea will be useful provided efficiently catch them and resist to hydrodynamic pressure. Nowadays design and installation of algae barrages are empiric. Structural break, anchor and mooring chain drifts can arise. We provide a mechanical model to evaluate stresses and loads on a structure made of nets and moorings. Hydrodynamic uncertainties occur through catches, fouling, and a range of sea currents appearing in lagoons or sheltered bays. We consider a nonlinear finite-element model of elastic membrane in large displacements and small strains. We study some forcings fixed a priori on a standard section of barrage. A section 50 m long, 1 m high, with two net-faces is examined at low and moderate current velocities, 0.05 and 0.35 m.s-1, while assuming specific vertical and horizontal catch pressures. A barrage installed in lagoon at Le François on Martinique Island and being observable by satellite imagery could benefit of the computed net and mooring tensions.

In Holter monitoring, the precise detection of standard heartbeats and Ventricular Premature Contractions (PVCs) is paramount for accurate cardiac rhythm assessment. This study introduces a novel application of the 1D U-Net neural network architecture aimed at enhancing PVC detection in Holter recordings. Training data comprised the Icentia11k, INCART DB, and our custom dataset. The model’s efficacy was subsequently validated against traditional Holter analysis methodologies across multiple databases, including AHA DB, MIT 11 DB, NST, and another custom dataset encompassing challenging real-world examples. The results underscored the 1D U-Net model’s prowess in QRS complex detection, achieving near-perfect balanced accuracy scores across all databases. PVC detection exhibited variability, with balanced accuracy scores ranging from 0.909 to 0.986. Despite some databases, like the AHA DB, showcasing lower sensitivity metrics, their robust balanced accuracy accentuates the model’s equitable performance in discerning both false positives and negatives. In conclusion, while the 1D U-Net architecture is a formidable tool for QRS detection, there’s a clear avenue for further refinement in its PVC detection capability, given the inherent complexities and noise challenges in real-world PVC occurrences.

The energy efficiency of existing buildings is a crucial factor in addressing Europe’s energy con-sumption challenges, accounting for nearly 40% of the total energy usage. The ongoing energy crisis, driven by the reliance on fossil fuels and increasing prices, has underscored the urgency of adopting sustainable and energy efficient practices especially to the existing building stock. Cyprus is par-ticularly vulnerable due to its heavy dependence on imported fuels for electricity generation and faces significant challenges in transforming its existing building stock into energy-efficient and sustainable structures. Extensive focus has been made by the government on the energy-efficient retrofit of non-residential public buildings. This study examines an existing cultural hall. The case study of the Cultural Hall represents the building stock before 2010 which is lacking thermal in-sulation, double glazing and any energy efficient element. It then offers an ideal case study to ex-plore sustainable solutions which can be easily adopted from the construction industry and from the building owners. The primary objective of this research is twofold: first, to develop a comprehensive retrofit plan to transform the cultural hall into a net-zero energy building; and second, to explore how this transformation can give value back to the local community. After running two simulation scenarios and providing two energy performance certificates, one with at the existing state and one with the intended retrofit measures, it is clear that the building can be a net zero energy consump-tion building. Some interesting results derived, present also how this building can provide value back to the community i.e. through solar carports for EV charging.

In this paper, we redefine the boundaries of blood cell classification, expanding from 5 to 11 distinct blood cell types using the challenging 11-class PBC dataset. This shift enables a deeper understanding of blood cell diversity, overcoming previous limitations in medical image analysis. Our approach combines state-of-the-art deep learning techniques, including pre-trained ConvNets, ViTb16 models, and custom CNN architectures. We employ transfer learning, fine-tuning, and ensemble strategies, such as CBAM and Averaging ensembles, to achieve unprecedented accuracy and interpretability. Our fully fine-tuned EfficientNetV2 B0 model sets a new standard, with a macro-average precision, recall, and F1-score of 91%, 90%, and 90%, respectively, and an average accuracy of 93%. This breakthrough underscores the transformative potential of 11-class blood cell classification for more precise medical diagnoses. Moreover, our groundbreaking "Naturalize" augmentation technique produces remarkable results. The 2K-PBC dataset generated with "Naturalize" boasts a macro-average precision, recall, and F1-score of 97%, along with an average accuracy of 96% when leveraging the fully fine-tuned EfficientNetV2 B0 model. This innovation not only elevates classification performance but also addresses data scarcity and bias in medical deep learning. Our research marks a paradigm shift in blood cell classification, enabling more nuanced and insightful medical analyses. The "Naturalize" technique’s impact extends beyond blood cell classification, emphasizing the vital role of diverse and comprehensive datasets in advancing healthcare applications through deep learning.

We investigated whether cigarette smoking and dietary acid load (DAL) are associated with the risk of chronic obstructive pulmonary disease (COPD) in middle-aged healthy Korean men. Healthy men without diagnosed chronic disease (40-64 yrs) from the KNHANES-VI (2013–2015) included in the analysis (n=774) were subdivided by smoking status and DAL levels as estimated by the net endogenous acid production (NEAP) quartile. Current smokers tended to have a higher risk of COPD than non-smokers before and after the adjustment. When divided by the DAL quartile, the Q4 tended to have a higher risk of COPD than the Q1. Additionally, current smokers with lower (Q2), modest (Q3), and highest NEAP scores (Q4) showed more than 4-fold higher risks of COPD than the non-smokers with the lowest NEAP scores (Q1). Ex–smokers with higher NEAP scores (Q3 and Q4) showed more than 4-fold higher risks of COPD than the Q1. Interestingly, the risk of COPD was also more than 6-fold higher in non-smokers with the highest NEAP scores compared to the Q1. NEAP scores and smoking status synergistically increased the risk of COPD in middle-aged healthy Korean men. It suggests that DAL levels are an important factor for the prevention and management of COPD.

A real-time monitoring system for the operational status of fishing vessels is an essential element for the modernization of the fishing industry. The operational status of fishing vessels can be identified by using onboard cameras to detect the deck crew and the use of fishing nets. Due to the typically limited processing capacity of shipboard equipment and the significant memory consumption of detection models, however, general target detection models are unable to perform real-time image detection to identify the operational status of fishing vessels. In this paper, we propose a lightweight real-time deck crew and the use of fishing nets detection method, YOLOv5s-SGC. It is based on the YOLOv5s model, which uses surveillance cameras to obtain video of fishing vessels operating at sea and enhances the dataset. YOLOv5s-SGC replaces YOLOv5s’s backbone and the ordinary convolutional blocks in the feature fusion network with ShufflNetV2 and Ghost module.

Vein segmentation and projection correction constitute the core algorithms of auxiliary venipuncture device, responding to the accurate venous positioning to assist puncture and reduce the number of punctures and pain of patients. This paper proposes an improved U-Net for segmenting vein and a coaxial correction for image alignment in the self-build vein projection system. The proposed U-Net is embedded by Gabor convolution kernels in the shallow layers to enhance segmentation accuracy. Additionally, to mitigate the semantic information loss caused by channel reduction, the network model is lightweighted by the mean of replacing conventional convolutions with inverted residual blocks. During the visualization process, a method that combining coaxial correction and homography matrix is proposed to address the non-planarity of the dorsal hand in this paper. First, use a hot mirror to adjust the light paths of both projector and camera to be coaxial, and then align the projected image with dorsal hand using a homography matrix. Using this approach, the device requires only a single calibration before use. With the implementation of the improved segmentation method, an accuracy rate of 95.12% is achieved by the dataset. The intersection over union ratio between the segmented and original images is reached at 90.07%. The entire segmentation process is completed in a 0.09 second, and the largest distance error of vein projection onto dorsal hand is 0.53mm. Experiments show that the device has reached practical accuracy and has the value of research and application.

In order to achieve global climate goals, it is necessary to estimate greenhouse gases (GHG) fluxes from ecosystems. To obtain a comprehensive assessment of CO2, СН4, N2O natural fluxes for the Russian Federation we used the "bottom-up" method according to [1]. We updated estimates for forest ecosystems based on State Forest Inventory data and satellite monitoring of forest disturbances. For grassland ecosystems, it was based on the correct distribution of areas between steppe and non-steppe zones. The estimated net uptake of natural ecosystems of Russia was 1.1 ± 1.8 billion tons of CO2-eq./year. The study shows that if only CO2 is taken into account, the net absorption of terrestrial ecosystems in Russia corresponds to more than -2.5 billion tons of CO2 (35% of forests’ contribution). However, given the emissions of non-CO2 GHGs, total net absorption in Russia’s natural ecosystems is reduced to about -1 billion tons of CO2-eq (with the forests’ contribution increasing to 80%). With regard to anthropogenic fluxes, the overall balance of GHGs in Russia corresponds to net-emissions of 1 billion tons of CO2-eq/year into the atmosphere. To improve reporting under the Paris Agreement, countries should aim to include only anthropogenic ("manageable") GHG fluxes on managed land.

Owing to the growing concerns due to dwindling fossil fuel reserves, increasing energy demand and climate emergency, it is imperative to develop and deploy sustainable energy technologies to ensure future energy supply as well as transition to the net zero world. In this context, there is great potential in the biorefinery concept for supplying drop in biofuels in the form of biodiesel. Biodiesel as a fuel can certainly bridge the gap where electrification or the use of hydrogen is not feasible, for instance heavy vehicles, farm and marine transportation sector. However, biodiesel industry also leads to formation of large amount of crude glycerol as the byproduct. Crude glycerol due to presence of several impurities may not be a suitable feedstock for all high value products derived from glycerol but fits well with glycerol esterification for producing glycerol acetins, which have numerous applications. This review critically looks at the processes using nanostructured solid/liquid acid catalysts for glycerol esterification including the economic viability of the scale up.

Quantification of the trade-offs among greenhouse gas (GHG) emissions, yield and farmers’ incomes is essential for proposing economic and environment nitrogen (N) management strategies for optimizing agricultural production. A four-year (2017–2020) field experiment (including four treatments: control experiment (CK), suitable utilization of fertilization (SU), emission reduction treatment (ER), and high fertilization (HF)) was conducted on maize (Zea mays L.) in the North China Plain. The Life Cycle Assessment (LCA) method was used in this study to quantify the GHG emissions and farmers’ incomes during the whole maize production process. The total GHG emissions of CK, SU, ER and HF treatments in the process of maize production are 10,755.2, 12,908.7, 11,950.1, and 14,274.5 kg CO2-eq ha−1 respectively, of which the direct emissions account for 84.8%, 76.8%, 74.9%, and 71.0%, respectively. Adding inhibitor significantly reduced direct GHG emissions, and the N2O and CO2 emissions from the maize fields in the ER treatment decreased by 30.0% and 7.9% compared to those in the SU treatment. Insignificant differences in yield were found between the SU and ER treatments, indicating that adding fertilizer inhibitors did not affect farmers’ incomes while reducing GHG emissions. The yield for SU, ER and HF treatments all significantly increased by 12.9%–24.0%, 10.0%–20.7% and 2.1%–17.4% compared to CK, respectively. In comparison with CK, both SU and ER significantly promoted agricultural net profit (ANP) by 16.6% and 12.2% with the mean ANP values of 3,101.0 USD ha-1 and 2,980.0 USD ha-1, respectively. Due to the high agricultural inputs, the ANP values in the HF treatment was 11.2%, 16.6% and 12.4% lower than those in the SU treatment in 2018–2020. In conclusion, the combination of N fertilizer and inhibitors was proved to be an environmentally friendly, high-profit and low-emissions production technology while sustaining or even increasing maize yields in the North China Plain, which was conducive to achieving agricultural sustainability.

(1) Background: Learning and training in variable conditions favors adapting to unstable or changing environments. The aim of the study was to test the effect of variable practice on the accuracy of the forehand net approach shot in tennis.; (2) Methods: Thirty (n=30) first-class national players (12.9 ± 1.1 years old) participated, divided into three groups: i) induced variability training (n=10) (varying court surfaces and balls), ii) specific training (n=10) and iii) usual training -control group- (n=10). All groups trained for a month: 12 sessions of 20 minutes -3 per week-. The accuracy of the shots were analyzed through a 2D capture and digitization process of the ball bounce on the court; (3) Results: The variability group presented better accuracy values after the period without practice than the stable training group (p = 0.041; ES= 0.51).; (4) Conclusions: the application of variability in the game conditions during tennis training seems to have a favorable effect on the retention of accuracy in the forehand down the line approach to the net.

Conducting computational stress-strain analysis using finite element methods (FEM) is a common approach when dealing with the complex geometries of atherosclerosis, which is a leading cause of global mortality and complex cardiovascular disease. The considerable expense linked to FEM analysis encourages the substitution of FEM with a considerably faster data-driven machine learning (ML) approach. This study investigated the potential of end-to-end deep learning tools as a more effective substitute for FEM in predicting stress-strain fields within 2D cross sections of arterial wall. We first proposed a U-Net based fully convolutional neural network (CNN) to predict the von Mises stress and strain distribution based on the spatial arrangement of calcification within arterial wall cross-sections. Further, we developed a conditional generative adversarial network (cGAN) to enhance, particularly from the perceptual perspective, the prediction accuracy of stress and strain field maps for arterial walls with various calcification quantities and spatial configurations. On top of U-Net and cGAN, we also proposed their ensemble approaches to further improve the prediction accuracy of field maps. Our dataset, consisting of input and output images, was generated by implementing boundary conditions and extracting stress-strain field maps. The trained U-Net models can accurately predict von Mises stress and strain fields, with structural similarity index scores (SSIM) of 0.854 and 0.830 and mean squared errors of 0.017 and 0.018 for stress and strain, respectively, on a reserved test set. Meanwhile, the cGAN models in a combination of ensemble and transfer learning techniques demonstrate high accuracy in predicting von Mises stress and strain fields, as evidenced by SSIM scores of 0.890 for stress and 0.803 for strain. Additionally, mean squared errors of 0.008 for stress and 0.017 for strain further support the model's performance on a designated test set. Overall, this study developed a surrogate model for finite element analysis, which can accurately and efficiently predict stress-strain fields of arterial walls regardless of complex geometries and boundary conditions.

Deep learning (DL) algorithms used for DOTATATE PET lesion detection typically require large, well-annotated training datasets. These are difficult to obtain due to low incidence of gastroenteopancreatic neuroendocrine tumors (GEP-NETs), and the high cost of manual annotation. Furthermore, networks trained and tested with data acquired from site specific PET/CT instrumentation, acquisition and processing protocols have reduced performance when tested with offsite data. This lack of generalizability requires even larger, more diverse training datasets. The objective of this study is to investigate the feasibility of improving DL algorithm performance by better matching the background noise in training datasets to higher noise, out-of-domain testing datasets. 68Ga-DOTATATE PET/CT datasets were obtained from two scanners: Scanner1, a state-of-the-art digital PET/CT (GE DMI PET/CT; n=83 subjects), and Scanner2, an older-generation analog PET/CT (GE STE; n=123 subjects). Set1, the data set from Scanner1 was reconstructed with standard clinical parameters (5 minutes; Q.Clear) and list-mode reconstructions (VPFXS 2, 3, 4, and 5-minutes). Set2, data from Scanner2, representing out-of-domain clinical scans, used standard iterative reconstruction (5 minutes; OSEM). Reconstructed datasets were divided into training, validation, and testing datasets in a 60%, 20%, 20% proportion, and a deep neural network was trained with each: Network1 for Scanner1 and Network2 for Scanner2. DL performance (Network1) was tested with out-of-domain test data (Set2). To evaluate the effect of training sample size, we tested DL model performance using a fraction (25%, 50%, and 75%) of Set1 for training. Scanner1, list-mode 2-minute reconstructed data demonstrated the most similar noise level compared that of Set2, resulting in the best performance (F1=0.713). This was not significantly different compared to the highest performance, upper-bound limit using in-domain training for Network2 (F1=0.755; p-value=0.103). Regarding sample size, the F1 score significantly increased from 25% training data (F1=0.478) to 100% training data (F1=0.713; p<0.001). List-mode data from modern PET scanners can be reconstructed to better match the noise properties of older scanners. Using existing data and their associated annotations dramatically reduces the cost and effort in generating these datasets, and significantly improves the performance of existing DL algorithms. List-mode reconstructions can provide an efficient, low-cost method to improve DL algorithm generalizability.

Rapidly and accurately extracting tobacco plant information can facilitate tobacco planting man-agement, precise fertilization and yield prediction. In karst plateau of southern China, tobacco plant identification is affected by large ground undulations, fragmented planting areas, complex and diverse habitats and uneven plant growth. This study took a tobacco planting area in Gui-zhou Province as the research object and used DJI UAVs to collect UAV visible light images. Considering plot fragmentation, plant size, presence of weeds and shadow masking, this area was classified into eight habitats. The datasets of different habitats were constructed to train the U-Net model. The results show that (1) the overall precision, recall, F1-score and IOU of tobacco plant information extraction were 0.68, 0.85, 0.75 and 0.60, respectively. (2) The precision was the highest for the subsurface-fragmented and weed-free habitat and the lowest for the smooth-tectonics and weed-infested habitat. (3) The weed-infested habitat with smaller tobacco plants can cause blurred images, reducing the plant identification accuracy. This study verified the feasibility of the U-Net model for tobacco single-plant identification in complex habitats. Decomposing complex habitats to establish the sample set method is a new attempt to improve crop identification in complex habitats in karst mountainous areas.

Color image binarization plays a pivotal role in image preprocessing work, significantly impacting subsequent tasks, particularly in text recognition. This paper concentrates on Document Image Binarization (DIB), aiming to separate a image into foreground (text) and background (non-text content). Through a thorough analysis of conventional and deep learning-based approaches, we conclude that prevailing DIB methods leverage deep learning technology. Furthermore, we explore the receptive fields pre- and post-network training to underscore the Transformer model's advantages. Subsequently, we introduce a lightweight model based on the U-Net structure, enhanced with the Mobile ViT module to better capture global information features in document images. Given its adeptness at learning both local and global features, our proposed model exhibits superior performance on two standard datasets (DIBCO2012 and DIBCO2017) compared to state-of-the-art methods. Notably, our proposed DIB method presents a straightforward end-to-end model devoid of additional image preprocessing or post-processing. Moreover, its parameter count is less than a quarter of the HIP'23 model, which achives best results on three datasets(DIBCO2012, DIBCO2017 and DIBCO2018). Finally, two sets of ablation experiments were conducted to verify the effectiveness of the proposed binarization model.

In areas like commercial and policy decisions—it's important to know how different treatments affect things and how well they work. However, treatments affect things on average—but now—with fancy computer methods and big sets of data, interest lies in figuring out how treatments might work differently for different people. For instance, let's say instead of just looking at what works for most people, researchers want to know which treatment is best for each person. This would help doctors to decide the right treatment for each patient. This idea isn't just for healthcare—it can help in other areas too, like deciding how to govern employees or making government policies. There are some ways people have been trying to do this already, but they struggle when there are lots of different factors to consider. New computer methods, though, are good at handling lots of factors, but they need a ton of data to work well. Therefore, this study looks at three computer methods that are—Virtual Twin Random Forest (VTRF), Causal Forest (CF), and Causal Net (CN) using real-world data about encouraging people to do manual labour. One problem this study is trying to solve is something called the "winner's curse", where a treatment might seem better than it really is because of how it's predicted by the computer. We are also testing some ideas to see if we can fix this problem and make the predictions more accurate.

The impact of the agriculture net carbon effect is crucial for climate change mitigation, yet research on this topic in China is currently insufficient. In this study, the carbon balance ratio (CBR) is used to characterise the net carbon effect of agriculture, and the spatial Durbin model and threshold regression model were used to empirically analyse the CBR with a sample of 30 provincial-level regions in China from 2004 to 2019. The aim was to investigate the mechanism of the influence of the land management scale (SCALE) of farming on the CBR. The results of the study indicate that national farming generally has a net sink effect, with large changes in interannual fluctuations; the expansion of SCALE has an obvious role in increasing sinks and reducing emissions, and has a positive spatial spillover effect on CBR, with significant spatial heterogeneity. Furthermore, as the SCALE expands, the direction of the influence of rural residents’ income and education level on CBR changes in the opposite direction, showing a significant non-linear relationship characteristic. However, in most areas, the SCALE threshold has not yet been reached, suggesting that the improvement in education levels can still offset the negative impact of economic income growth in the short term. The study recommends paying attention to the role of farmers’ human capital, changing production methods, and implementing carbon peaking and carbon neutrality goals to enhance the net sink effect of the farming industry, which would be achieved by increasing land management scale.

A single paragraph of about 200 words maximum. For research articles, abstracts should give a pertinent overview of the work. We strongly encourage authors to use the following style of structured abstracts, but without headings: (1) Background: Place the question addressed in a broad context and highlight the purpose of the study; (2) Methods: briefly describe the main methods or treatments applied; (3) Results: summarize the article’s main findings; (4) Conclusions: indicate the main conclusions or interpretations. The abstract should be an objective representation of the article and it must not contain results that are not presented and substantiated in the main text and should not exaggerate the main conclusions.

To enable net zero sustainable thermal building energy this study develops an open-source thermal house model to couple solar photovoltaic (PV) and heat pumps (HP) for grid-connected residential housing. The calculation of both space heating and cooling thermal loads and the selection of HP is accomplished with a validated Python model for air-source heat pumps. The capacity of PV required to supply the HPs is calculated using an System Advisor Model integrated Python model. Self-sufficiency and self-consumption of PV and the energy imported/exported to the grid for a case study are provided, which shows that simulations based on the monthly load profile have a significant reduction of 43% for energy sent to/from the grid compared to the detailed hourly simulation and an increase from 30% to 60% for self-consumption and self-sufficiency. These results show the importance of more granular modeling and also indicate mismatches of PV generation and HP load based on hourly simulation datasets. The back-calculation PV sizing algorithm combined with HP and thermal loads presented in this study exhibited robust performance. The results indicate this approach can be used to accelerate the solar electrification of heating and cooling to offset the use of fossil fuels in northern climates.