Convolutional Block Research Articles

Reduction of Vision-Based Models for Fall Detection

Due to the limitations that falls have on humans, early detection of these becomes essential to avoid further damage. In many applications, various technologies are used to acquire accurate information from individuals such as wearable sensors, environmental sensors or cameras, but all of these require high computational resources in many cases, delaying the response of the entire system. The complexity of the models used to process the input data and detect these activities makes them almost impossible to complete on devices with limited resources, which are the ones that could offer an immediate response avoiding unnecessary communications between sensors and centralized computing centers. In this work, we chose to reduce the models to detect falls using images as input data. We proceeded to use image sequences as video frames, using data from two open source datasets, and we applied the Sparse Low Rank Method to reduce certain layers of the Convolutional Neural Networks that were the backbone of the models. Additionally, we chose to replace a convolutional block with Long Short Term Memory to consider the latest updates of these data sequences. The results showed that performance was maintained decently while significantly reducing the parameter size of the resulting models.

RJ-TinyViT: an efficient vision transformer for red jujube defect classification.

Compared to the surface defect detection of industrial products produced according to specified processes, the detection of surface defects in naturally grown red jujubes poses unique and significant challenges for researchers. The high diversity of surface defects, subtle distinctions from the background, low contrast, varying scales, and the presence of high levels of noise in images are among the factors that greatly amplify the complexity of defect detection tasks. Existing methods show some deficiencies in addressing these issues, mainly due to insufficient feature extraction capabilities and overly complex network structures, leading to limitations in model efficiency and practical application performance. To tackle the challenges associated with red jujube surface defect detection, this study proposes an optimized Tiny Vision Transformer (TinyViT) network structure, named RJ-TinyViT. This method refines the TinyViT-5m network structure to reduce network burden and introduces an improved Multi-Kernel Block (MK Block) and an improved Mobile Inverted Bottleneck Convolution Block (MBConv Block) to enhance feature extraction capabilities. Additionally, we have integrated the Coordinate Attention (CA) module to enhance the model's capacity for recognizing and focusing on features of surface defects on red jujubes. Experimental results show that RJ-TinyViT achieved a classification accuracy of 93.38%, marking an improvement of 1.84% over the original TinyViT network. At the same time, its Floating-point Operations (FLOPs) and Parameters (Params) were reduced to 58.97% and 39.84% of the original TinyViT network, respectively. These results not only demonstrate that RJ-TinyViT achieves model lightweighting while maintaining high accuracy but also highlight its value in practical industrial applications.

A Hybrid Deep Learning Model for Enhanced Structural Damage Detection: Integrating ResNet50, GoogLeNet, and Attention Mechanisms

Quick and accurate structural damage detection is essential for maintaining the safety and integrity of infrastructure, especially following natural disasters. Traditional methods of damage assessment, which rely on manual inspections, can be labor-intensive and subject to human error. This paper introduces a hybrid deep learning model that combines the capabilities of ResNet50 and GoogLeNet, further enhanced by a convolutional block attention module (CBAM), proposed to improve both the accuracy and performance in detecting structural damage. For training purposes, a diverse dataset of images depicting both structural damage cases and undamaged cases was used. To further enhance the robustness, data augmentation techniques were also employed. In this research, precision, recall, F1-score, and accuracy were employed to evaluate the effectiveness of the introduced hybrid deep learning model. Our findings indicate that the hybrid deep neural network introduced in this study significantly outperformed standalone architectures such as ResNet50 and GoogLeNet, making it a highly effective solution for applications in disaster response and infrastructure maintenance.

KOC_Net: Impact of the Synthetic Minority Over-Sampling Technique with Deep Learning Models for Classification of Knee Osteoarthritis Using Kellgren–Lawrence X-Ray Grade

One of the most common diseases afflicting humans is knee osteoarthritis (KOA). KOA occurs when the knee joint cartilage breaks down, and knee bones start rubbing together. The diagnosis of KOA is a lengthy process, and missed diagnosis can have serious consequences. Therefore, the diagnosis of KOA at an initial stage is crucial which prevents the patients from Severe complications. KOA identification using deep learning (DL) algorithms has gained popularity during the past few years. By applying knee X-ray images and the Kellgren–Lawrence (KL) grading system, the objective of this study was to develop a DL model for detecting KOA. This study proposes a novel model based on CNN called knee osteoarthritis classification network (KOC_Net). The KOC_Net model contains 05 convolutional blocks, and each convolutional block has three components such as Convlotuioanl2D, ReLU, and MaxPooling 2D. The KOC_Net model is evaluated on two publicly available benchmark datasets which consist of X-ray images of KOA based on the KL grading system. Additionally, we applied contrast-limited adaptive histogram equalization (CLAHE) methods to enhance the contrast of the images and utilized SMOTE Tomek to deal with the problem of minority classes. For the diagnosis of KOA, the classification performance of the proposed KOC_Net model is compared with baseline deep networks, namely Dense Net-169, Vgg-19, Xception, and Inception-V3. The proposed KOC_Net was able to classify KOA into 5 distinct groups (including Moderate, Minimal, Severe, Doubtful, and Healthy), with an AUC of 96.71%, accuracy of 96.51%, recall of 91.95%, precision of 90.25%, and F1-Score of 96.70%. Dense Net-169, Vgg-19, Xception, and Inception-V3 have relative accuracy rates of 84.97%, 81.08%, 87.06%, and 83.62%. As demonstrated by the results, the KOC_Net model provides great assistance to orthopedics in making diagnoses of KOA.

A Lightweight Barcode Detection Algorithm Based on Deep Learning

For existing situations of missed detections, false detections, and repeated detections in barcode detection algorithms in real-world scenarios, a barcode detection algorithm based on improved YOLOv8 is proposed. The EfficientViT block based on a linear self-attention mechanism is introduced into the backbone of the original model to enhance the model’s attention to barcode features. In the model’s neck, linear mapping and grouped convolution are used to improve the C2f module, and the ADown convolution block is utilized to modify the model’s downsampling, which reduces the model’s parameters and computational cost while improving the efficiency of model feature fusion. Finally, the reconstruction of the model’s detection head and the modification of the loss function are implemented to enhance the model’s training quality and reduce the model’s error in barcode detection. Experimental results indicate that the improved model exhibits an increase of 1.8% in recall rate and 1.9% in mAP50:95 for barcode localization and classification. The FPS is improved by 40 frames per second. The model’s parameter count is reduced by 74.2%, and FLOPs are decreased by 79.6%. Furthermore, the proposed model outperforms other models in terms of model size and barcode detection accuracy.

Cucumber Leaf Segmentation Based on Bilayer Convolutional Network

When monitoring crop growth using top-down images of the plant canopies, leaves in agricultural fields appear very dense and significantly overlap each other. Moreover, the image can be affected by external conditions such as background environment and light intensity, impacting the effectiveness of image segmentation. To address the challenge of segmenting dense and overlapping plant leaves under natural lighting conditions, this study employed a Bilayer Convolutional Network (BCNet) method for accurate leaf segmentation across various lighting environments. The major contributions of this study are as follows: (1) Utilized Fully Convolutional Object Detection (FCOS) for plant leaf detection, incorporating ResNet-50 with the Convolutional Block Attention Module (CBAM) and Feature Pyramid Network (FPN) to enhance Region of Interest (RoI) feature extraction from canopy top-view images. (2) Extracted the sub-region of the RoI based on the position of the detection box, using this region as input for the BCNet, ensuring precise segmentation. (3) Utilized instance segmentation of canopy top-view images using BCNet, improving segmentation accuracy. (4) Applied the Varifocal Loss Function to improve the classification loss function in FCOS, leading to better performance metrics. The experimental results on cucumber canopy top-view images captured in glass greenhouse and plastic greenhouse environments show that our method is highly effective. For cucumber leaves at different growth stages and under various lighting conditions, the Precision, Recall and Average Precision (AP) metrics for object recognition are 97%, 94% and 96.57%, respectively. For instance segmentation, the Precision, Recall and Average Precision (AP) metrics are 87%, 83% and 84.71%, respectively. Our algorithm outperforms commonly used deep learning algorithms such as Faster R-CNN, Mask R-CNN, YOLOv4 and PANet, showcasing its superior capability in complex agricultural settings. The results of this study demonstrate the potential of our method for accurate recognition and segmentation of highly overlapping leaves in diverse agricultural environments, significantly contributing to the application of deep learning algorithms in smart agriculture.

Improved Prototypical Network Model for Classification of Farmland Shelterbelt Using Sentinel-2 Imagery

Farmland shelterbelt plays an important role in protecting farmland and ensuring stable crop yields, and it is mainly distributed in the form of bands and patches; different forms of distribution have different impacts on farmland, which is an important factor affecting crop yields. Therefore, high-precision classification of banded and patch farmland shelterbelt is a prerequisite for analyzing its impact on crop yield. In this study, we explored the effectiveness and transferability of an improved Prototypical Network model incorporating data augmentation and a convolutional block attention module for extracting banded and patch farmland shelterbelt in Northeast China, and we analyzed the potential of applying it to the production of large-scale farmland shelterbelt products. Firstly, we classified banded and patch farmland shelterbelt under different sample window sizes using the improved Prototypical Network in the source domain study area to obtain the optimal sample window size and the optimal classification model. Secondly, fine-tuning transfer learning and learning from scratch directly were used to classify the banded and patch farmland shelterbelt in the target domain study area, respectively, to evaluate the extraction model’s migratability. The results showed that classification of farmland shelterbelt using the improved Prototypical Network is very effective, with the highest extraction accuracy under the 5 × 5 sample window; the accuracies of the banded and patch farmland shelterbelt are 92.16% and 90.91%, respectively. Using the fine-tuning transfer learning method in the target domain can classify the banded and patch farmland shelterbelt with high accuracy, above 95% and 89%, respectively. The proposed approach can provide new insight into farmland shelterbelt classification and farmland shelterbelt products obtained from freely accessible Sentinel-2 multispectral images.

A Deep Learning Model for Accurate Maize Disease Detection Based on State-Space Attention and Feature Fusion

In improving agricultural yields and ensuring food security, precise detection of maize leaf diseases is of great importance. Traditional disease detection methods show limited performance in complex environments, making it challenging to meet the demands for precise detection in modern agriculture. This paper proposes a maize leaf disease detection model based on a state-space attention mechanism, aiming to effectively utilize the spatiotemporal characteristics of maize leaf diseases to achieve efficient and accurate detection. The model introduces a state-space attention mechanism combined with a multi-scale feature fusion module to capture the spatial distribution and dynamic development of maize diseases. In experimental comparisons, the proposed model demonstrates superior performance in the task of maize disease detection, achieving a precision, recall, accuracy, and F1 score of 0.94. Compared with baseline models such as AlexNet, GoogLeNet, ResNet, EfficientNet, and ViT, the proposed method achieves a precision of 0.95, with the other metrics also reaching 0.94, showing significant improvement. Additionally, ablation experiments verify the impact of different attention mechanisms and loss functions on model performance. The standard self-attention model achieved a precision, recall, accuracy, and F1 score of 0.74, 0.70, 0.72, and 0.72, respectively. The Convolutional Block Attention Module (CBAM) showed a precision of 0.87, recall of 0.83, accuracy of 0.85, and F1 score of 0.85, while the state-space attention module achieved a precision of 0.95, with the other metrics also at 0.94. In terms of loss functions, cross-entropy loss showed a precision, recall, accuracy, and F1 score of 0.69, 0.65, 0.67, and 0.67, respectively. Focal loss showed a precision of 0.83, recall of 0.80, accuracy of 0.81, and F1 score of 0.81. State-space loss demonstrated the best performance in these experiments, achieving a precision of 0.95, with recall, accuracy, and F1 score all at 0.94. These results indicate that the model based on the state-space attention mechanism achieves higher detection accuracy and better generalization ability in the task of maize leaf disease detection, effectively improving the accuracy and efficiency of disease recognition and providing strong technical support for the early diagnosis and management of maize diseases. Future work will focus on further optimizing the model’s spatiotemporal feature modeling capabilities and exploring multi-modal data fusion to enhance the model’s application in real agricultural scenarios.

A Novel Defect Segmentation Model via Integrating Enhanced U-Net with Convolutional Block Attention Mechanisms

Abstract. Defect segmentation in industrial quality control is pivotal for maintaining high product surface standards, but it remains a challenge due to complex defect patterns and varying backgrounds. Traditional segmentation methods often struggle with sensitivity and specificity, leading to either missed defects or false positives. In order to address the above issues this paper presents an advanced defect segmentation model that integrates a Convolutional Block Attention Module (CBAM) with an enhanced U-Net architecture. Our method leverages the spatial and channel-wise attention mechanisms to focus more accurately on subtle defects and remove background interference, enhancing the discriminative power of the feature maps generated by the U-Net. Experimental results on SD-saliency-900 industrial datasets demonstrate significant improvements in segmentation accuracy and robustness compared to existing state-of-the-art methods, confirming the effectiveness of integrating attention mechanisms in deep learning models for complex defect segmentation tasks. Specifically, our model achieved a mean Intersection over Union (mIoU) of 0.87, which is an increase of 5% over the traditional U-Net model and a 4% improvement compared to the DeepLabv3+ model. These results indicate a substantial enhancement in defect detection performance, validating the superiority of our approach.

FusionGCN: Multi-focus image fusion using superpixel features generation GCN and pixel-level feature reconstruction CNN

In recent years, convolutional neural networks have demonstrated significant advancements in the domain of computer vision, effectively addressing numerous previously challenging issues. An increasing number of researchers are focusing their investigations on this field, proposing innovative network architectures. However, many existing networks necessitate intricate module designs and a substantial number of parameters to achieve satisfactory fusion outcomes, which poses challenges for lightweight devices with constrained computational resources. To mitigate this concern, the present study introduces a novel methodology that integrates block segmentation with pixel optimization. Specifically, we initially employ graph convolutional networks to execute flexible convolutions on large-scale, irregular regions generated through superpixel clustering, thereby achieving coarse segmentation at the block level. Subsequently, we utilize parallel lightweight convolutional networks to provide pixel-level guidance, ultimately resulting in a more accurate decision map. Furthermore, to leverage the strengths of both networks and facilitate the optimization of feature generation from the graph convolutional network for non-Euclidean data, we meticulously design a superpixel-based graph decoder alongside a pixel-based convolutional neural network extraction block to enhance feature acquisition and propagation. In comparison to numerous state-of-the-art methodologies, our approach demonstrates commendable competitiveness in both qualitative and quantitative analyses, as well as in efficiency evaluations. The code can be downloaded at https://github.com/ouyangbaicai/FusionGCN.

DeepLigType: Predicting Ligand Types of Protein-Ligand Binding Sites Using a Deep Learning Model.

The analysis of protein-ligand binding sites plays a crucial role in the initial stages of drug discovery. Accurately predicting the ligand types that are likely to bind to protein-ligand binding sites enables more informed decision making in drug design. Our study, DeepLigType, determines protein-ligand binding sites using Fpocket and then predicts the ligand type of these pockets with the deep learning model, Convolutional Block Attention Module (CBAM) with ResNet. CBAM-ResNet has been trained to accurately predict five distinct ligand types. We classified protein-ligand binding sites into five different categories according to the type of response ligands cause when they bind to their target proteins, which are antagonist, agonist, activator, inhibitor, and others. We created a novel dataset, referred to as LigType5, from the widely recognized PDBbind and scPDB dataset for training and testing our model. While the literature mostly focuses on the specificity and characteristic analysis of protein binding sites by experimental (laboratory-based) methods, we propose a computational method with the DeepLigType architecture. DeepLigType demonstrated an accuracy of 74.30% and an AUC of 0.83 in ligand type prediction on a novel test dataset using the CBAM-ResNet deep learning model. For access to the code implementation of this research, please visit our GitHub repository at https://github.com/drorhunvural/DeepLigType.

BreakNet: discontinuity-resilient multi-scale transformer segmentation of retinal layers

Visible light optical coherence tomography (vis-OCT) is gaining traction for retinal imaging due to its high resolution and functional capabilities. However, the significant absorption of hemoglobin in the visible light range leads to pronounced shadow artifacts from retinal blood vessels, posing challenges for accurate layer segmentation. In this study, we present BreakNet, a multi-scale Transformer-based segmentation model designed to address boundary discontinuities caused by these shadow artifacts. BreakNet utilizes hierarchical Transformer and convolutional blocks to extract multi-scale global and local feature maps, capturing essential contextual, textural, and edge characteristics. The model incorporates decoder blocks that expand pathways to enhance the extraction of fine details and semantic information, ensuring precise segmentation. Evaluated on rodent retinal images acquired with prototype vis-OCT, BreakNet demonstrated superior performance over state-of-the-art segmentation models, such as TCCT-BP and U-Net, even when faced with limited-quality ground truth data. Our findings indicate that BreakNet has the potential to significantly improve retinal quantification and analysis.

Combination of edge enhancement and cold diffusion model for low dose CT image denoising.

Since the quality of low dose CT (LDCT) images is often severely affected by noise and artifacts, it is very important to maintain high quality CT images while effectively reducing the radiation dose. In recent years, the representation of diffusion models to produce high quality images and stable trainability has attracted wide attention. With the extension of the cold diffusion model to the classical diffusion model, its application has greater flexibility. Inspired by the cold diffusion model, we proposes a low dose CT image denoising method, called CECDM, based on the combination of edge enhancement and cold diffusion model. The LDCT image is taken as the end point (forward) of the diffusion process and the starting point (reverse) of the sampling process. Improved sobel operator and Convolution Block Attention Module are added to the network, and compound loss function is adopted. The experimental results show that CECDM can effectively remove noise and artifacts from LDCT images while the inference time of a single image is reduced to 0.41 s. Compared with the existing LDCT image post-processing methods, CECDM has a significant improvement in all indexes.

A Lightweight and Efficient Multimodal Feature Fusion Network for Bearing Fault Diagnosis in Industrial Applications

To address the issues of single-structured feature input channels, insufficient feature learning capabilities in noisy environments, and large model parameter sizes in intelligent diagnostic models for mechanical equipment, a lightweight and efficient multimodal feature fusion convolutional neural network (LEMFN) method is proposed. Compared with existing models, LEMFN captures rich fault features at multiple scales by combining time-domain and frequency-domain signals, thereby enhancing the model’s robustness to noise and improving data adaptability under varying operating conditions. Additionally, the convolutional block attention module (CBAM) and random overlapping sampling technology (ROST) are introduced, and through a feature fusion strategy, the accurate diagnosis of mechanical equipment faults is achieved. Experimental results demonstrate that the proposed method not only possesses high diagnostic accuracy and rapid convergence but also exhibits strong robustness in noisy environments. Finally, a graphical user interface (GUI)-based mechanical equipment fault detection system was developed to promote the practical application of intelligent fault diagnosis in mechanical equipment.

Enhancing the Rainfall Forecasting Accuracy of Ensemble Numerical Prediction Systems via Convolutional Neural Networks

Abstract Ensemble prediction systems are commonly used to demonstrate the potential uncertainty of weather forecasts. These systems also help provide weather predictions to prepare for disasters. Specifically, a type of prediction called a quantitative precipitation forecast (QPF) is calculated from the average of multiple forecasts. The probability-matched mean (PMM) method is used to improve these QPF predictions when they underestimate rainfall. However, the PMM method can exhibit limitations when dealing with certain types of rain patterns. To address this issue, this study focuses on improving short-term heavy rainfall predictions using an artificial intelligence (AI) algorithm that combines deep neural networks (DNNs), including convolutional neural networks (CNNs), with a space-based attention mechanism (convolutional block attention module [CBAM]). Four experiments were conducted to evaluate the new method, including those performed to optimize the timing of a 24-hour QPF model, adjust the training dataset, and refine the training algorithm. Two specific weather events were assessed: rainfall influenced by southwest winds after typhoons and afternoon thunderstorm rainfall. A comparison of the results obtained via the root mean square error (RMSE) and structural similarity index measure (SSIM) reveals that the accuracy and distribution of the QPF predictions significantly improved over those of the PMM method. Compared with the PMM method, our approach can reduce the RMSE from 77.51 to 19.52 in the Mei-yu case, an improvement of approximately 74.82%. The SSIM also increased from 0.33 to 0.56, indicating a 70.9% enhancement. Overall, the new approach successfully enhances rainfall prediction accuracy via AI techniques and has the potential to be applied in disaster preparedness operations.

Quantitative characterization of surface defects on bridge cable based on improved YOLACT++

The safety and reliability of cables are directly linked to the safe operation of bridges as crucial load-bearing components. The accuracy and efficiency of current methods are still insufficient to segment and quantitatively characterize surface defects on cables. This paper proposes a novel and efficient method for the refined segmentation and quantitative characterization of bridge cable surface defects based on an improved you only look at coefficients++ (YOLACT++) model. For defect segmentation, several enhancements have been made to the YOLACT++ model, including incorporating the convolutional block attention module (CBAM), optimizing the anchor box generation mechanism, and introducing the smoother Mish activation function, which enhances both the accuracy and speed of defect detection. For quantitative characterization, the method adopts surface correction algorithms, pixel statistics, and crack skeleton extraction, resulting in a more accurate representation of defect areas and the length and width of cracks. Compared to the baseline model, the optimized model achieves a 3.58 % improvement in mean average precision (mAP) and an inference speed of 25.74 frames per second (FPS). The results show that the error is within 10 % compared with the manually measured area, which offers a more objective and comprehensive foundation for cable safety assessment.

Feature Enhancement Based Oriented Object Detection in Remote Sensing Images

Since objects in remote sensing imagery often have arbitrary orientations and high densities, the features of small objects are inclined to be contaminated by the background and other instances. To address the issues, we propose a new oriented object detection framework where a series of feature enhancement schemes are implemented so as to improve robustness and accuracy of the detector. Firstly, we design a weighted bidirectional feature pyramid network, which can be used to fuse both high-level semantic features and low-level detail features for effectively handling with multi-scale objects. Accordingly, we apply the convolutional block attention module that exploits both spatial- and channel-wise attention in our detector, and study how to effectively integrate it into the framework for adaptive feature refinement. In the meanwhile, we present a semantic segmentation guided module to generate naive mask, which is used to multiple with pyramid features to filter out background noise and improve feature representation for small objects. The experimental results on two public datasets, i.e., UCAS-AOD and DOTA, validate the effective performance of the proposed method for oriented object detection in remote sensing images.

A method for enhancing near-mirror object detection by integrating AWCS and CBAM into ResNet18

ABSTRACT Defects, such as scratches, frequently arise during the manufacturing of near-mirror workpieces, adversely affecting both their longevity and aesthetic appeal. The identification of these near-mirror defects through machine vision is fraught with challenges, including the scarcity of comprehensive datasets, variability in surface curvature, and the diminutive size of the defects. In response to these challenges, this study integrates Visual Parser and deep learning methodologies for the visual detection of near-mirror defects. A dataset conducive to deep learning was developed utilising a sliding window approach. The proposed model incorporates an Adaptive Weighted Convolution Structure (AWCS) and Convolutional Block Attention Module (CBAM) within the ResNet18 architecture. Additionally, we introduce an efficient and precise technique termed Visual Parser, which facilitates the detection and localisation of defect areas in near-mirror workpieces. When combined with a classification model, the Visual Parser allows for the online inference and detection of extensive near-mirror images. Experimental findings indicate that the performance metrics of the proposed ResNet18-AWCS-CBAM (ResNet18-ASCM) model significantly exceed those of established networks. The defect detection and localisation method based on Visual Parser is capable of fulfilling the online inspection requirements pertinent to the production of near-mirror workpieces.

Adaptation of Object Detection Algorithms for Road Indicator Lights in Complex Scenes

In the realm of autonomous driving, practical driving scenarios are fraught with numerous complexities, including inclement weather conditions, nighttime blurriness, and ambient light sources that significantly hinder drivers’ ability to discern road indicators. Furthermore, the dynamic nature of road indicators, which are constantly evolving, poses additional challenges for computer vision-based detection systems. To address these issues, this paper introduces a road indicator light detection model, leveraging the advanced capabilities of YOLOv8. We have ingeniously integrated the robust backbone of YOLOv8 with four distinct attention mechanism modules—Convolutional Block Attention Module (CBAM), Efficient Channel Attention (ECA), Shuffle Attention (SA), and Global Attention Mechanism (GAM)—to significantly enhance the model performance in capturing nuanced features of road indicators and boosting the accuracy of detecting minute objects. Additionally, we employ the Asymptotic Feature Pyramid Network (AFPN) strategy, which optimizes the fusion of features across different scales, ensuring not only an enhanced performance but also maintaining real-time capability. These innovative attention modules empower the model by recalibrating the significance of both channel and spatial dimensions within the feature maps, enabling it to hone in on the most pertinent object characteristics. To tackle the challenges posed by samples rich in small, occluded, background-similar objects, and those that are inherently difficult to recognize, we have incorporated the Focaler-IOU loss function. This loss function deftly reduces the contribution of easily detectable samples to the overall loss, thereby intensifying the focus on challenging samples. This strategic balancing of hard-to-detect versus easy-to-detect samples effectively elevates the model’s detection performance. Experimental evaluations conducted on both a public traffic signal dataset and a proprietary headlight dataset have yielded impressive results, with both mAP50 and mAP50:95 metrics experiencing significant improvements exceeding two percentage points. Notably, the enhancements observed in the headlight dataset are particularly profound, signifying a significant step forward toward realizing safer and more reliable assisted driving technologies.

FINE VESSEL SEGMENTATION WITH REFINEMENT GATE IN DEEP LEARNING ARCHITECTURES

Automated vessel segmentation is essential in diagnosing eye-related disorders and monitoring progressive retinal diseases. State-of-the-art methods have achieved excellent results in this field, but very few have considered the post-processing of feature maps. As a result, there is often a lack of small and fine vessels or discontinuities in segmented vessels. To address this issue, this study introduces a novel post-processing technique called the refinement gate, which works with a deep learning model during training. The refinement gate enhances contextual information to extract important features from feature maps better. The proposed technique is applied with U-net architecture and placed after every convolution block in the encoder path. Visual and statistical comparisons demonstrate the robustness of the proposed method using three publicly available datasets, namely: the DRIVE DB, the STARE DB, and CHASE_DB1 datasets, showing significant improvements to segment weak and tiny vessels. The reported results confirm the potential of the model to be used as a segmentation tool in the medical field. This study is the first to propose such a gating mechanism without additional trainable parameters or standalone networks as in other literature.