Convolutional Block Research Articles

M2ANet: Multi-branch and multi-scale attention network for medical image segmentation

Abstract Convolutional neural networks (CNNs)-based medical image segmentation technologies have been widely used in medical image segmentation because of their strong representation and generalization abilities. However, due to the inability to effectively capture global information from images, CNNs can easily lead to loss of contours and textures in segmentation results. Notice that the transformer model can effectively capture the properties of long-range dependencies in the image, and furthermore, combining the CNN and the transformer can effectively extract local details and global contextual features of the image. Motivated by this, we propose a multi-branch and multi-scale attention network (M2ANet) for medical image segmentation, whose architecture consists of three components. Specifically, in the first component, we construct an adaptive multi-branch patch module for parallel extraction of image features to reduce information loss caused by downsampling. In the second component, we apply residual block to the well-known convolutional block attention module to enhance the network’s ability to recognize important features of images and alleviate the phenomenon of gradient vanishing. In the third component, we design a multi-scale feature fusion module, in which we adopt adaptive average pooling and position encoding to enhance contextual features, and then multi-head attention is introduced to further enrich feature representation. Finally, we validate the effectiveness and feasibility of the proposed M2ANet method through comparative experiments on four benchmark medical image segmentation datasets, particularly in the context of preserving contours and textures. The source code of M2ANet will be released at https://github.com/AHUT-MILAGroup/M2ANet.

A Multi-Scale Fusion Convolutional Network for Time-Series Silicon Prediction in Blast Furnaces

In steel production, the blast furnace is a critical element. In this process, precisely controlling the temperature of the molten iron is indispensable for attaining efficient operations and high-grade products. This temperature is often indirectly reflected by the silicon content in the hot metal. However, due to the dynamic nature and inherent delays of the ironmaking process, real-time prediction of silicon content remains a significant challenge, and traditional methods often suffer from insufficient prediction accuracy. This study presents a novel Multi-Scale Fusion Convolutional Neural Network (MSF-CNN) to accurately predict the silicon content during the blast furnace smelting process, addressing the limitations of existing data-driven approaches. The proposed MSF-CNN model extracts temporal features at two distinct scales. The first scale utilizes a Convolutional Block Attention Module, which captures local temporal dependencies by focusing on the most relevant features across adjacent time steps. The second scale employs a Multi-Head Self-Attention mechanism to model long-term temporal dependencies, overcoming the inherent delay issues in the blast furnace process. By combining these two scales, the model effectively captures both short-term and long-term temporal dependencies, thereby enhancing prediction accuracy and real-time applicability. Validation using real blast furnace data demonstrates that MSF-CNN outperforms recurrent neural network models such as Long Short-Term Memory (LSTM) and the Gated Recurrent Unit (GRU). Compared with LSTM and the GRU, MSF-CNN reduces the Root Mean Square Error (RMSE) by approximately 22% and 21%, respectively, and improves the Hit Rate (HR) by over 3.5% and 4%, highlighting its superiority in capturing complex temporal dependencies. These results indicate that the MSF-CNN adapts better to the blast furnace’s dynamic variations and inherent delays, achieving significant improvements in prediction precision and robustness compared to state-of-the-art recurrent models.

Fractal-Based Architectures with Skip Connections and Attention Mechanism for Improved Segmentation of MS Lesions in Cervical Spinal Cord

Background/Objectives: Multiple sclerosis (MS) is an autoimmune disease that damages the myelin sheath of the central nervous system, which includes the brain and spinal cord. Although MS lesions in the brain are more frequently investigated, MS lesions in the cervical spinal cord (CSC) can be much more specific for the diagnosis of the disease. Furthermore, as lesion burden in the CSC is directly related to disease progression, the presence of lesions in the CSC may help to differentiate MS from other neurological diseases. Methods: In this study, two novel deep learning models based on fractal architectures are proposed for the automatic detection and segmentation of MS lesions in the CSC by improving the convolutional and connection structures used in the layers of the U-Net architecture. In our previous study, we introduced the FractalSpiNet architecture by incorporating fractal convolutional block structures into the U-Net framework to develop a deeper network for segmenting MS lesions in the CPC. In this study, to improve the detection of smaller structures and finer details in the images, an attention mechanism is integrated into the FractalSpiNet architecture, resulting in the Att-FractalSpiNet model. In addition, in the second hybrid model, a fractal convolutional block is incorporated into the skip connection structure of the U-Net architecture, resulting in the development of the Con-FractalU-Net model. Results: Experimental studies were conducted using U-Net, FractalSpiNet, Con-FractalU-Net, and Att-FractalSpiNet architectures to detect the CSC region and the MS lesions within its boundaries. In segmenting the CSC region, the proposed Con-FractalU-Net architecture achieved the highest Dice Similarity Coefficient (DSC) score of 98.89%. Similarly, in detecting MS lesions within the CSC region, the Con-FractalU-Net model again achieved the best performance with a DSC score of 91.48%. Conclusions: For segmentation of the CSC region and detection of MS lesions, the proposed fractal-based Con-FractalU-Net and Att-FractalSpiNet architectures achieved higher scores than the baseline U-Net architecture, particularly in segmenting small and complex structures.

A Lightweight Dual-Branch Complex-Valued Neural Network for Automatic Modulation Classification of Communication Signals

Currently, deep learning has become a mainstream approach for automatic modulation classification (AMC) with its powerful feature extraction capability. Complex-valued neural networks (CVNNs) show unique advantages in the field of communication signal processing because of their ability to directly process complex data and obtain signal amplitude and phase information. However, existing models face deployment challenges due to excessive parameters and computational complexity. To address these limitations, a lightweight dual-branch complex-valued neural network (LDCVNN) is proposed. The framework uses dual pathways to separately capture features with phase information and complex-scaling-equivariant representations, adaptively fused via trainable weighted fusion. Spatial and channel reconstruction convolution (SCConv) is extended to complex domain and combined with complex-valued depthwise separable convolution block (CBlock) and complex-valued average pooling to eliminate feature redundancy and extract higher order features. Finally, efficient classification is realized through complex-valued fully connected layers and a complex-valued Softmax. The evaluations demonstrate that LDCVNN achieves the highest average accuracy on RML2016.10a with only 9.0 K parameters and without data augmentation, which reducing the number of parameters by 99.33% compared to CDSN and by 97.25% compared to CSDNN. Additionally, LDCVNN achieves a better balance between efficiency and performance across other datasets.

Anomaly detection in pneumatic control valves using stacked re-optimized convolutional autoencoder and product quantization under imbalanced data conditions

Abstract Reliable operation of pneumatic control valves is essential for the safety and efficiency of industrial control systems. Existing data-driven anomaly detection methods are typically based on datasets containing faults from faulty valves. However, the scarcity of abnormal samples and tedious or time-consuming of data labeling in practice often lead to imbalanced datasets and unlabeled data, which pose challenges for the application of these methods. To address this problem, we propose a novel anomaly detection method for pneumatic control valves based on stacked re-optimized convolutional autoencoder (S-RCAE) and product quantization (PQ) technology. S-RCAE is designed as a feature extraction network to obtain the latent and essential feature representations of anomalies in the monitoring data. It is stacked by placing a sample selection layer with the DBSCAN clustering algorithm between the two CAEs. Furthermore, the two CAEs are modified by inserting a convolutional block attention module between every two layers. During the training of S-RCAE, the idea of few-shot learning and an improved cost-sensitive function are adopted for the labeling and feature extraction of anomaly samples in unlabeled and class-imbalanced datasets. Then, a feature fusion mechanism is implemented for the resulting representation from S-RCAE. The PQ is further introduced to enable dimensionality reduction and storage of the fused features and the accurate identification and assessment of anomalies using quantitative scores. The experimental results show that the proposed method can accurately identify and quantify the abnormal states of the valve from the unlabeled and class-imbalanced datasets.

Sound Signal-based Lightweight Fault Diagnosis for Railway Turnout System via Combining Transformer and CNN

Abstract Significant progress has been made in deep learning-based fault diagnosis for railway turnout system. However, the increasing complexity of networks has resulted in higher network parameters and computational demands, making it necessary for algorithms to meet more stringent hardware requirements. Therefore, there is a need for a lightweight deep learning approach that can effectively diagnose faults in railway turnout system while maintaining high performance. In this research, we propose a lightweight model integrating Transformer and Convolutional Neural Networks (CNN) for fault diagnosis in railway turnout systems. The model combines CNN and Transformer to extract both local and global features from monitoring signals and introduces lightweight enhancements. Specifically, a Multi-Scale Depthwise Dilated Convolution (MDDC) block is introduced into the CNN to break down cross-channel convolutions into pointwise and depthwise convolutions, with multiscale dilated convolution kernels applied in the depthwise convolutions. Additionally, an Context-Aware Attention Module (CAAM) is integrated into the Transformer to simplify complex matrix multiplications and multidimensional exponential operations. Experimental results using sound signals collected during actuator operations demonstrate that the optimized algorithm reduces the number of parameters and computational complexity by 75% compared to the baseline model, while also reducing inference time by one-third. The optimized algorithm achieves an accuracy of 99.54%. These findings lay the groundwork for further optimization of fault diagnosis in railway turnout system.

Optimized Variational Mode Decomposition and Convolutional Block Attention Module-Enhanced Hybrid Network for Bearing Fault Diagnosis

Accurate fault diagnosis remains a critical but unresolved issue in predictive maintenance, as industrial environments typically involve large amounts of electromagnetic interference and mechanical noise that can severely degrade the signal quality. In this study, we propose an innovative diagnostic framework to address the challenging problem of bearing fault diagnosis in vibration signals under complex noise conditions. We develop the VMD-CNN-BiLSTM-CBAM model by systematically integrating the variational mode decomposition (VMD), convolutional neural network (CNN), bi-directional long and short-term memory network (BiLSTM), and convolutional block attention module (CBAM). The framework starts with VMD-based signal decomposition, which effectively separates the noise component from the bearing vibration features. Based on this denoising, a CNN architecture is employed to extract multi-scale spatio-temporal features through its hierarchical learning mechanism. The subsequent BiLSTM layer captures bidirectional temporal dependencies to model fault-evolution patterns, while the CBAM module strategically highlights key diagnostic features through adaptive channel spatial attention. Experimental validation using the Case Western Reserve University and Jiangnan University bearing datasets demonstrates the excellent performance of the model, with average accuracies of 99.76% and 99.40%, respectively. Finally, additional validation through our customized testbed confirms the usefulness of the model with an average accuracy of 99.70%. These results demonstrate that the proposed approach greatly improves fault diagnosis in noisy industrial environments through its synergistic architectural design and enhanced noise immunity.

A Reinforced, Event-Driven, and Attention-Based Convolution Spiking Neural Network for Multivariate Time Series Prediction

Despite spiking neural networks (SNNs) inherently exceling at processing time series due to their rich spatio-temporal information and efficient event-driven computing, the challenge of extracting complex correlations between variables in multivariate time series (MTS) remains to be addressed. This paper proposes a reinforced, event-driven, and attention-based convolution SNN model (REAT-CSNN) with three novel features. First, a joint Gramian Angular Field and Rate (GAFR) coding scheme is proposed to convert MTS into spike images, preserving the inherent features in MTS, such as the temporal patterns and spatio-temporal correlations between time series. Second, an advanced LIF-pooling strategy is developed, which is then theoretically and empirically proved to be effective in preserving more features from the regions of interest in spike images than average-pooling strategies. Third, a convolutional block attention mechanism (CBAM) is redesigned to support spike-based input, enhancing event-driven characteristics in weighting operations while maintaining outstanding capability to capture the information encoded in spike images. Experiments on multiple MTS data sets, such as stocks and PM2.5 data sets, demonstrate that our model rivals, and even surpasses, some CNN- and RNN-based techniques, with up to 3% better performance, while consuming significantly less energy.

Multi-Modality Sheep Face Recognition Based on Deep Learning

To address the challenge of recognizing sheep faces of the same type, which exhibit significant similarities and varying performance of RGB images under different lighting conditions and angles, this paper proposes a dual-branch multi-modal sheep face recognition model based on the ResNet18 architecture. This model effectively learns geometric features from depth data and texture features from RGB data, thereby enhancing recognition accuracy. Initially, the model employs two InceptionV2 layers, one for the RGB channel and another for the depth channel, to extract specific features from both modalities. Subsequently, the losses from the two modalities are computed. In the mid-stage, the two modalities are fused using the Convolutional Block Attention Module (CBAM), and in the final stage, a residual network is utilized to learn the complementary features between the modalities. Experimental results demonstrate that this model benefits from effective multi-modal fusion, achieving high accuracy in sheep face recognition under complex lighting conditions and various angles.

QuARF: Quality-Adaptive Receptive Fields for Degraded Image Perception

Advanced Deep Neural Networks (DNNs) perform well for high-quality images, but their performance dramatically decreases for degraded images. Data augmentation is commonly used to alleviate this problem, but using too much perturbed data might seriously decrease the performance on pristine images. To tackle this challenge, we take our cue from the assumption of spatial coincidence in human visual perception, i.e. multiscale and varying receptive fields are required for understanding pristine and degraded images. Correspondingly, we propose a novel plug-and-play network architecture, dubbed Quality-Adaptive Receptive Fields (QuARF), to automatically select the optimal receptive fields based on the quality of the input image. To this end, we first design a multi-kernel convolutional block, which comprises multiscale continuous receptive fields. Afterward, we design a quality-adaptive routing network to predict the significance of each kernel, based on the quality features extracted from the input image. In this way, QuARF automatically selects the optimal inference route for each image. To further boost efficiency and effectiveness, the input feature map is split into multiple groups, with each group independently learning its quality-adaptive routing parameters. We apply QuARF to a variety of DNNs and conduct experiments in both discriminative and generation tasks, including semantic segmentation, image translation, and restoration. Thorough experimental results show that QuARF significantly and robustly improves the performance for degraded images, and outperforms data augmentation in most cases.

Improved Lightweight YOLOv11 Algorithm for Real-Time Forest Fire Detection

Modern computer vision techniques for forest fire detection face a trade-off between computational efficiency and detection accuracy in complex forest environments. To address this, we propose a lightweight YOLOv11n-based framework optimized for edge deployment. The backbone network integrates a novel C3k2MBNV2 (Cross Stage Partial Bottleneck with 3 convolutions and kernel size 2 MobileNetV2) block to enable efficient fire feature extraction via a compact architecture. We further introduce the SCDown (Spatial-Channel Decoupled Downsampling) block in both the backbone and neck to preserve critical information during downsampling. The neck further incorporates the C3k2WTDC (Cross Stage Partial Bottleneck with 3 convolutions and kernel size 2, combined with Wavelet Transform Depthwise Convolution) block, enhancing contextual understanding with reduced computational overhead. Experiments on a forest fire dataset demonstrate that our model achieves a 53.2% reduction in parameters and 28.6% fewer FLOPs compared to YOLOv11n (You Only Look Once version eleven), along with a 3.3% improvement in mean average precision. These advancements establish an optimal balance between efficiency and accuracy, enabling the proposed framework to attain real-time detection capabilities on resource-constrained edge devices in forest environments. This work provides a practical solution for deploying reliable forest fire detection systems in scenarios demanding low latency and minimal computational resources.

Postoperative Recovery Assessment for Parkinson's Patients via Light-weighted Topological Pose Estimation.

The UPDRS III scale plays a critical role in diagnosing the progression of Parkinson's disease. Current methods often involve doctors guiding patients through specific actions on the scale, recording their performance, and assigning scores. However, this approach has several drawbacks, including the lengthy time required for doctorpatient communication, the high costs of patients traveling to hospitals for follow-up visits, and the reliance on subjective judgments from doctors, which lack standardized criteria. With advancements in artificial intelligence, many traditional processes have been partially automated. To help patients reduce diagnosis time, lower medical costs, and provide more accurate and objective evaluation results, this paper proposes a Transformer-based pose estimation model for assessing UPDRS III scale actions. By integrating skeleton-based evaluations from the network with a series of post-processing operations, the model enables patients to perform self-assessments of their post-treatment recovery at home, saving doctors significant time. This work introduces a cascaded graph self-attention module, SGAM (Spatial-Graphical Attention Module), to enhance the network's understanding of human topology. Additionally, it proposes a lightweight convolutional block, Chi-block, which employs a novel approach leveraging the attribute invariance of filters to interpret model performance and guide compression. This approach reduces computational costs and model parameters while preserving accuracy. The proposed method demonstrates robust performance on human pose estimation (HPE) datasets and showcases impressive lightweight performance on benchmark datasets such as ImageNet-1K and CIFAR-10. These results demonstrate the potential of artificial intelligence in enabling automated remote diagnosis and treatment for Parkinson's patients.

Optimizing Waste Management with Squeeze-and-Excitation and Convolutional Block Attention Integration in ResNet-Based Deep Learning Frameworks

Effective waste classification is crucial for sustainable waste management, yet existing automated models face challenges such as misclassification of visually similar waste materials, dataset imbalance, and poor generalization to real-world conditions. This study addresses these limitations by integrating Squeeze-and-Excitation (SE) and Convolutional Block Attention Module (CBAM) mechanisms into ResNet-50, a deep learning model, to improve waste classification accuracy. Utilizing the TrashNet dataset (six waste categories), we have employed the Synthetic Minority Over-sampling Technique (SMOTE) to balance the dataset, increasing samples from 2,527 to 3,564. Experimental results demonstrate significant improvements: ResNet-50 achieves 74.42% test accuracy, SE+ResNet-50 improves accuracy to 93.47%, and CBAM+ResNet-50 reaches 95.74%. The findings highlight that attention-based deep learning models can significantly enhance feature extraction, improve classification accuracy, and optimize waste segregation processes, contributing to more efficient recycling and waste management automation. Future work includes extending the model to classify hazardous and mixed waste, ensuring broader applicability in real-world waste management systems.

YOLOv5_CDB: A Global Wind Turbine Detection Framework Integrating CBAM and DBSCAN

Wind energy plays a crucial role in global sustainable development, and accurately estimating the number and spatial distribution of wind turbines is crucial for strategic planning and energy allocation. To address the critical need for wind turbine detection and spatial distribution analysis, this study develops YOLOv5_CDB, an enhanced detection framework based on the YOLOv5 model. The proposed method incorporates two key components: the Convolutional Block Attention Mechanism (CBAM) to improve feature representation and the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm for spatial density clustering. The method is applied to 2 m resolution World Imagery data. It detects both tubular and lattice wind turbines by analyzing key features, including turbine towers and shadows. The YOLOv5_CDB demonstrates a substantial enhancement in performance when compared with the YOLOv5s. The F1-score shows an increase of 1.39%, and the mean average precision (mAP) exhibits a 1.5% improvement. Meanwhile, the precision (P) and recall (R) values are recorded at 95.97% and 91.18%, respectively. Furthermore, YOLOv5_CDB evinces consistent performance advantages, outperforming state-of-the-art models including YOLOv8s, YOLOv12s, and RT-DETR by 1.84%, 3.98%, and 1.77% in terms of F1-score and by 3.7%, 4.5%, and 3.0% in terms of mAP, respectively. The YOLOv5_CDB model has been demonstrated to show superior performance in the global wind turbine detection domain, thereby providing a foundation for the management of wind farms and the development of sustainable energy.

Prediction of Extensibility and Toughness of Wheat-Flour Dough Using Bubble Inflation–Structured Light Scanning 3D Imaging Technology and the Enhanced 3D Vgg11 Model

The extensibility of dough and its resistance to extension (toughness) are important indicators, since they are directly linked to dough quality. Therefore, this paper used an independently developed device to blow sheeted dough, and then a three-dimensional (3D) camera was used to continuously collect point cloud images of sheeted dough forming bubbles. After data collection, the rotation algorithm, region of interest (ROI) extraction algorithm, and statistical filtering algorithm were used to process the original point cloud images. Lastly, the oriented bounding box (OBB) algorithm was proposed to calculate the deformation height of each data point. And the point cloud image with the largest deformation depth was selected as the data to input into the 3D convolutional neural network (CNN) models. The Convolutional Block Attention Module (CBAM) was introduced into the 3D Visual Geometry Group 11 (Vgg11) model to build the enhanced Vgg11. And we compared it with the other classical 3D CNN models (MobileNet, ResNet18, and Vgg11) by inputting the voxel-point-based data and the voxel-based data separately into these models. The results showed that the enhanced 3D Vgg11 model using voxel-point-based data was superior to the other models. For prediction of dough extensibility and toughness, the Rp was 0.893 and 0.878, respectively.

A Tractor Transmission Systems Vibration Fault Data Augmentation and Diagnosis Method Based on DCGAN

The actual production process is often faced with insufficient fault data, Traditional data augmentation algorithms are prone to problems such as overfitting and pattern collapse. To address the issues, a tractor transmission systems vibration fault data augmentation and diagnosis method based on deep convolutional generative adversarial network is proposed. First, the original vibration signal is converted to a time-frequency diagram, which serves as the input to the generative adversarial network. Subsequently, the Two Timescale Update Rule(TTUR) strategy and gradient penalty are applied to stabilize the training process of the generative adversarial network, and the self-attention mechanism is introduced to enhance the feature extraction capability of the generative adversarial network. Furthermore, an enhanced AlexNet based on the Convolutional Block Attention Module is developed to improve the diagnostic performance. The approach was verified using datasets from Case Western Reserve University (CWRU), Huazhong University of Science and Technology (HUST), and a laboratory gearbox dataset. Results show 99% diagnostic accuracy for CWRU, over 97.5% for the laboratory dataset, and 98% accuracy under a single condition for the HUST gear dataset, with accuracy exceeding 95.3% under other conditions. This indicates the method improves diagnostic precision and model generalization under data scarcity.

Neuromorphic imaging cytometry on human blood cells

Abstract Imaging flow cytometry (IFC) is a powerful cell analytic tool that exploits multi-parameters in single-cell images to characterise cell phenotypes and fluorescence information. It enables in-depth analysis of cell signalling, DNA repair and marker localisation. However, conventional frame-based acquisition is bound to the triangle of imaging constraints—speed, resolution and sensitivity, which has become an everlasting challenge to overcome during development. Neuromorphic photosensors detect contrast changes in a scene via individual-firing pixels, characterising superior data efficiency, temporal resolution and fluorescence sensitivity. In this work, we have developed a neuromorphic imaging cytometer (NIC) to capture fast-moving cell events, curating the first neuromorphic cell dataset with human blood cells, endothelial cells and artificial particles. Recently, this sensor has been adopted to address the limitations in IFC with prominent results in diverse modalities and machine learning approaches. Such a dataset serves as a baseline of healthy cell groups for both diagnostic and research purposes. In addition, the rich spatial information derived from cell images has exceptional uses with deep learning (DL) approaches to automate cell analysis, classification, sorting and gating strategy. We also trained a lightweight model combining the convolutional block attention module with a spiking neural network (CBAM-SNN) to automate cell analysis and classification. The proposed architecture has achieved a promising performance of 97% accuracy and F1 score with a significant reduction in computation requirements. Combining the data sparsity in neuromorphic imaging with a lightweight DL model and operation platform can enable next-generation, AI-driven cytometry to deliver point-of-care diagnostic and research solutions.

CNDD-Net: A Lightweight Attention-Based Convolutional Neural Network for Classifying Corn Nutritional Deficiencies and Leaf Diseases

Plant diseases and nutrient deficiencies pose significant challenges to food production, making it crucial to identify them accurately and quickly, as their symptoms can often be similar. Prompt and precise detection is essential to implement effective measures that prevent crop losses. While computer vision techniques have demonstrated effectiveness in classification, their high computational demands have limited their adoption by farmers in the field. In this study, a Corn leaf Nutrition Deficiency and Disease network (CNDD-net) is designed based on the ResNet framework, incorporating a depth-wise separable convolution and a convolutional block attention module for a lightweight, high-performance model. The models underwent training, validation, and testing using a corn leaf nutrition deficiencies and diseases data set with seven classes implementing five-fold cross-validation. The performance of the models is assessed using average accuracy, GFLOPs, number of parameters, and model size. Following experiments involving the manipulation of the position of the attention module, the number of feature maps, and the depth of the network, the model was finalised. The CNDD-net design has a model size of 0.24 MB with 48,041 parameters and a GFLOPs of 0.18, providing an average accuracy of 96.71%. Compared to conventional models, this research demonstrates optimal performance and computational complexity, offering an efficient, lightweight solution to identify nutritional deficiencies and diseases of corn leaf, thus supporting sustainable agriculture.

YOLOv8-Based Road Sign Detection Algorithm Incorporating CBAM and DWConv

Detecting road signs is essential for autonomous driving technology, especially in the identification of small objects.To overcome the difficulties of identifying tiny road signs, In order to increase detection performance, this work proposes an enhanced YOLOv8 method that combines Depthwise Separable Convolution (DWConv) with the Convolutional Block Attention Module (CBAM). Specifically, YOLOv8 serves as the baseline model, which is optimized in the feature extraction, fusion, and detection stages. The CBAM attention mechanism is incorporated into the Neck section, while traditional convolutions are replaced with DWConv, improving the model's focus on tiny information while reducing computational complexity. To improve the model's generalization ability, data augmentation methods like Mosaic and Mixup are incorporated. Mosaic augmentation increases the diversity of training data by stitching different images together, whereas Mixup improves the models adaptability to various scenes by blending images. Additionally, common augmentation techniques, including cropping, color adjustment, and flipping, are effectively applied to optimize model performance. Experimental results indicate that, compared with YOLOv8n, the improved YOLOv8 algorithm achieves a 2.1 percentage point increase in mean Average Precision (mAP0.5), a 4.9% improvement in mAP50-95, and a 7.2% increase in recall rate. Furthermore, the algorithm significantly reduces the missed detection rate and improves small-object detection performance while lowering runtime by 4.1%. These results demonstrate the practical applicability of the proposed method.

Segmentation detection method in tree-shaded environment for road cracks collected by inspection vehicle on WFU-Unet

Road cracks pose a significant safety hazard to transportation, making timely detection crucial for traffic safety. Traditional crack segmentation methods face three main issues: (1) Tree shadow background affects crack recognition in real-world environments. (2) Conventional convolutional neural networks fail to detect complete cracks. (3) Direct deconvolution during upsampling results in unclear crack details. To address these challenges, this paper proposes the WFU-Unet model for road crack detection and segmentation. First, the WCM module, constructed with wavelet transform, ConvNext, and MobileNet, reduces shadow interference, enabling the network to distinguish between cracks and tree shadows. Second, the Fuse module replaces traditional convolutional blocks, enhancing the network’s ability to extract crack features. Finally, the Up module substitutes conventional upsampling techniques to minimize spatial information loss of cracks. Experimental results show that the WFU-Unet model achieves a Miou of 81.68%, precision of 91.43%, recall of 86.63%, and F1-Score of 88.93%. Compared to other models, WFU-Unet demonstrates superior generalization ability and segmentation accuracy, making it more suitable for crack detection in shadowed environments.