Search Results (535)

Medical image segmentation plays a crucial role in medical image processing, focusing on the automated extraction of regions of interest (such as organs, lesions, etc.) from medical images. This process supports various clinical applications, including diagnosis, surgical planning, and treatment. In this paper, we introduce a Boundary-guided Context Fusion U-Net (BCF-UNet), a novel approach designed to tackle a critical shortcoming in current methods: the inability to effectively integrate boundary information with semantic context. The BCF-UNet introduces a Adaptive Multi-Frequency Encoder (AMFE), which uses multi-frequency analysis inspired by the Wavelet Transform (WT) to capture both local and global features efficiently. The Adaptive Multi-Frequency Encoder (AMFE) decomposes images into different frequency components and adapts more effectively to boundary texture information through a learnable activation function. Additionally, we introduce a new multi-scale feature fusion module, the Atten-kernel Adaptive Fusion Module (AKAFM), designed to integrate deep semantic information with shallow texture details, significantly bridging the gap between features at different scales. Furthermore, each layer of the encoder sub-network integrates a Boundary-aware Pyramid Module (BAPM), which utilizes a simple and effective method and combines it with a priori knowledge to extract multi-scale edge features to improve the accuracy of boundary segmentation. In BCF-UNet, semantic context is used to guide edge information extraction, enabling the model to more effectively comprehend and identify relationships among various organizational structures. Comprehensive experimental evaluations on two datasets demonstrate that the proposed BCF-UNet achieves superior performance compared to existing state-of-the-art methods. Full article

Botnets pose a significant challenge in network security by leveraging Domain Generation Algorithms (DGA) to evade traditional security measures. Extracting DGA domain samples is inherently complex, and the current DGA detection models often struggle to capture domain features effectively when facing limited training data. This limitation results in suboptimal detection performance and an imbalance between model accuracy and complexity. To address these challenges, this paper introduces a novel multi-scale feature fusion model that integrates the Transformer architecture with the Rapid Selective Kernel Network (R-SKNet). The proposed model employs the Transformer’s encoder to couple the single-domain character elements with the multiple types of relationships within the global domain block. This paper proposes integrating R-SKNet into DGA detection and developing an efficient channel attention (ECA) module. By enhancing the branch information guidance in the SKNet architecture, the approach achieves adaptive receptive field selection, multi-scale feature capture, and lightweight yet efficient multi-scale convolution. Moreover, the improved Feature Pyramid Network (FPN) architecture, termed EFAM, is utilized to adjust channel weights for outputs at different stages of the backbone network, leading to achieving multi-scale feature fusion. Experimental results demonstrate that, in tasks with limited training samples, the proposed method achieves lower computational complexity and higher detection accuracy compared to mainstream detection models. Full article

Aiming at many challenges in the recognition task of peach leaf shrink disease, such as the diversity of object size of diseased leaf disease, complex background interference, and inflexible adjustment of model training learning rate, we propose a peach leaf shrink disease recognition algorithm based on an attention generalized efficient layer aggregation network. Firstly, the rectified linear unit activation function is used to effectively improve the stability and performance of the model in low-precision computing environments and solve the problem of partial gradient disappearance. Secondly, the integrated squeeze-and-excitation network attention mechanism can adaptively focus on the key areas of pests and diseases in the image, which significantly enhances the recognition ability of the model to the characteristics of pests and diseases. Finally, combined with fast pyramid pooling enhanced with Local Attention Networks, the deep fusion of cross-layer features is realized to improve the ability of the model to identify complex features and optimize the operation efficiency. The experimental results on the peach leaf shrink disease recognition dataset show that the proposed algorithm achieves a significant improvement in performance compared with the original YOLOv8 algorithm. Specifically, mF1, mPrecision, mRecall, and mAP increased by 0.1075, 0.0723, 0.1224, and 0.1184, respectively, which provided strong technical support for intelligent and automatic monitoring of peach pests and diseases. Full article

Book localization is crucial for the development of intelligent book inventory systems, where the high-precision detection of book spines is a critical requirement. However, the varying tilt angles and diverse aspect ratios of books on library shelves often reduce the effectiveness of conventional object detection algorithms. To address these challenges, this study proposes an enhanced oriented R-CNN algorithm for book spine detection. First, we replace the standard 3 × 3 convolutions in ResNet50’s residual blocks with deformable convolutions to enhance the network’s capacity for modeling the geometric deformations of book spines. Additionally, the PAFPN (Path Aggregation Feature Pyramid Network) was integrated into the neck structure to enhance multi-scale feature fusion. To further optimize the anchor box design, we introduce an adaptive initial cluster center selection method for K-median clustering. This allows for a more accurate computation of anchor box aspect ratios that are better aligned with the book spine dataset, enhancing the model’s training performance. We conducted comparison experiments between the proposed model and other state-of-the-art models on the book spine dataset, and the results demonstrate that the proposed approach reaches an mAP of 90.22%, which outperforms the baseline algorithm by 4.47 percentage points. Our method significantly improves detection accuracy, making it highly effective for identifying book spines in real-world library environments. Full article

Grasping objects of irregular shapes and various sizes remains a key challenge in the field of robotic grasping. This paper proposes a novel RGB-D data-based grasping pose prediction network, termed Cascaded Feature Fusion Grasping Network (CFFGN), designed for high-efficiency, lightweight, and rapid grasping pose estimation. The network employs innovative structural designs, including depth-wise separable convolutions to reduce parameters and enhance computational efficiency; convolutional block attention modules to augment the model’s ability to focus on key features; multi-scale dilated convolution to expand the receptive field and capture multi-scale information; and bidirectional feature pyramid modules to achieve effective fusion and information flow of features at different levels. In tests on the Cornell dataset, our network achieved grasping pose prediction at a speed of 66.7 frames per second, with accuracy rates of 98.6% and 96.9% for image-wise and object-wise splits, respectively. The experimental results show that our method achieves high-speed processing while maintaining high accuracy. In real-world robotic grasping experiments, our method also proved to be effective, achieving an average grasping success rate of 95.6% on a robot equipped with parallel grippers. Full article

In order to achieve infrared aircraft detection under interference conditions, this paper proposes an infrared aircraft detection algorithm based on high-resolution feature-enhanced semantic segmentation network. Firstly, the designed location attention mechanism is utilized to enhance the current-level feature map by obtaining correlation weights between pixels at different positions. Then, it is fused with the high-level feature map rich in semantic features to construct a location attention feature fusion network, thereby enhancing the representation capability of target features. Secondly, based on the idea of using dilated convolutions to expand the receptive field of feature maps, a hybrid atrous spatial pyramid pooling module is designed. By utilizing a serial structure of dilated convolutions with small dilation rates, this module addresses the issue of feature information loss when expanding the receptive field through dilated spatial pyramid pooling. It captures the contextual information of the target, further enhancing the target features. Finally, a dice loss function is introduced to calculate the overlap between the predicted results and the ground truth labels, facilitating deep excavation of foreground information for comprehensive learning of samples. This paper constructs an infrared aircraft detection algorithm based on a high-resolution feature-enhanced semantic segmentation network which combines the location attention feature fusion network, the hybrid atrous spatial pyramid pooling module, the dice loss function, and a network that maintains the resolution of feature maps. Experiments conducted on a self-built infrared dataset show that the proposed algorithm achieves a mean intersection over union (mIoU) of 92.74%, a mean pixel accuracy (mPA) of 96.34%, and a mean recall (MR) of 96.19%, all of which outperform classic segmentation algorithms such as DeepLabv3+, Segformer, HRNetv2, and DDRNet. This demonstrates that the proposed algorithm can achieve effective detection of infrared aircraft in the presence of interference. Full article

To satisfy the obstacle avoidance requirements of unmanned agricultural machinery during autonomous operation and address the challenge of rapid obstacle detection in complex field environments, an improved field obstacle detection model based on YOLOv8 was proposed. This model enabled the fast detection and recognition of obstacles such as people, tractors, and electric power pylons in the field. This detection model was built upon the YOLOv8 architecture with three main improvements. First, to adapt to different tasks and complex environments in the field, improve the sensitivity of the detector to various target sizes and positions, and enhance detection accuracy, the CBAM (Convolutional Block Attention Module) was integrated into the backbone layer of the benchmark model. Secondly, a BiFPN (Bi-directional Feature Pyramid Network) architecture took the place of the original PANet to enhance the fusion of features across multiple scales, thereby increasing the model’s capacity to distinguish between the background and obstacles. Third, WIoU v3 (Wise Intersection over Union v3) optimized the target boundary loss function, assigning greater focus to medium-quality anchor boxes and enhancing the detector’s overall performance. A dataset comprising 5963 images of people, electric power pylons, telegraph poles, tractors, and harvesters in a farmland environment was constructed. The training set comprised 4771 images, while the validation and test sets each consisted of 596 images. The results from the experiments indicated that the enhanced model attained precision, recall, and average precision scores of 85.5%, 75.1%, and 82.5%, respectively, on the custom dataset. This reflected increases of 1.3, 1.2, and 1.9 percentage points when compared to the baseline YOLOv8 model. Furthermore, the model reached 52 detection frames per second, thereby significantly enhancing the detection performance for common obstacles in the field. The model enhanced by the previously mentioned techniques guarantees a high level of detection accuracy while meeting the criteria for real-time obstacle identification in unmanned agricultural equipment during fieldwork. Full article

Aiming at the problem of high false detection and missed detection rate of apple surface defects in complex environments, a new apple surface defect detection network: space-to-depth convolution-Multi-scale Empty Attention-Context Guided Feature Pyramid Network-You Only Look Once version 8 nano (SMC-YOLOv8n) is designed. Firstly, space-to-depth convolution (SPD-Conv) is introduced before each Faster Implementation of CSP Bottleneck with 2 convolutions (C2f) in the backbone network as a preprocessing step to improve the quality of input data. Secondly, the Bottleneck in C2f is removed in the neck, and Multi-scale Empty Attention (MSDA) is introduced to enhance the feature extraction ability. Finally, the Context Guided Feature Pyramid Network (CGFPN) is used to replace the Concat method of the neck for feature fusion, thereby improving the expression ability of the features. Compared with the YOLOv8n baseline network, mean Average Precision (mAP) 50 increased by 2.7% and 1.1%, respectively, and mAP50-95 increased by 4.1% and 2.7%, respectively, on the visible light apple surface defect data set and public data set in the self-made complex environments.The experimental results show that SMC-YOLOv8n shows higher efficiency in apple defect detection, which lays a solid foundation for intelligent picking and grading of apples. Full article

Background: Given the severe economic burden that citrus greening disease imposes on fruit farmers and related industries, rapid and accurate disease detection is particularly crucial. This not only effectively curbs the spread of the disease, but also significantly reduces reliance on manual detection within extensive citrus planting areas. Objective: In response to this challenge, and to address the issues posed by resource-constrained platforms and complex backgrounds, this paper designs and proposes a novel method for the recognition and localization of citrus greening disease, named the HHS-RT-DETR model. The goal of this model is to achieve precise detection and localization of the disease while maintaining efficiency. Methods: Based on the RT-DETR-r18 model, the following improvements are made: the HS-FPN (high-level screening-feature pyramid network) is used to improve the feature fusion and feature selection part of the RT-DETR model, and the filtered feature information is merged with the high-level features by filtering out the low-level features, so as to enhance the feature selection ability and multi-level feature fusion ability of the model. In the feature fusion and feature selection sections, the HWD (hybrid wavelet-directional filter banks) downsampling operator is introduced to prevent the loss of effective information in the channel and reduce the computational complexity of the model. Through using the ShapeIoU loss function to enable the model to focus on the shape and scale of the bounding box itself, the prediction of the bounding box of the model will be more accurate. Conclusions and Results: This study has successfully developed an improved HHS-RT-DETR model which exhibits efficiency and accuracy on resource-constrained platforms and offers significant advantages for the automatic detection of citrus greening disease. Experimental results show that the improved model, when compared to the RT-DETR-r18 baseline model, has achieved significant improvements in several key performance metrics: the precision increased by 7.9%, the frame rate increased by 4 frames per second (f/s), the recall rose by 9.9%, and the average accuracy also increased by 7.5%, while the number of model parameters reduced by ${0.137\times 10}^7$. Moreover, the improved model has demonstrated outstanding robustness in detecting occluded leaves within complex backgrounds. This provides strong technical support for the early detection and timely control of citrus greening disease. Additionally, the improved model has showcased advanced detection capabilities on the PASCAL VOC dataset. Discussions: Future research plans include expanding the dataset to encompass a broader range of citrus species and different stages of citrus greening disease. In addition, the plans involve incorporating leaf images under various lighting conditions and different weather scenarios to enhance the model’s generalization capabilities, ensuring the accurate localization and identification of citrus greening disease in diverse complex environments. Lastly, the integration of the improved model into an unmanned aerial vehicle (UAV) system is envisioned to enable the real-time, regional-level precise localization of citrus greening disease. Full article

Infrared small target detection technology has been widely applied in the defense sector, including applications such as precision targeting, alert systems, and naval monitoring. However, due to the small size of their targets and the extended imaging distance, accurately detecting drone targets in complex infrared environments remains a considerable challenge. Detecting drone targets accurately in complex infrared environments poses a substantial challenge. This paper introduces a novel model that integrates edge characteristics with multi-scale feature fusion, named Edge-Guided Feature Pyramid Networks (EG-FPNs). This model aims to capture deep image features while simultaneously emphasizing edge characteristics. The goal is to resolve the problem of missing target information that occurs when Feature Pyramid Networks (FPNs) perform continuous down-sampling to obtain deeper semantic features. Firstly, an improved residual block structure is proposed, integrating multi-scale convolutional feature extraction and inter-channel attention mechanisms, with significant features being emphasized through channel recalibration. Then, a layered feature fusion module is introduced to strengthen the shallow details in images while fusing multi-scale image features, thereby strengthening the shallow edge features. Finally, an edge self-fusion module is proposed to enhance the model’s depiction of image features by extracting edge information and integrating it with multi-scale features. We conducted comparative experiments on multiple datasets using the proposed algorithm and existing advanced methods. The results show improvements in the IoU, nIoU, and F1 metrics, while also showcasing the lightweight nature of EG-FPNs, confirming that they are more suitable for drone detection in resource-constrained infrared scenarios. Full article

Walnuts possess significant nutritional and economic value. Fast and accurate sorting of shells and kernels will enhance the efficiency of automated production. Therefore, we propose a FastQAFPN-YOLOv8s object detection network to achieve rapid and precise detection of unsorted materials. The method uses lightweight Pconv (Partial Convolution) operators to build the FasterNextBlock structure, which serves as the backbone feature extractor for the Fasternet feature extraction network. The ECIoU loss function, combining EIoU (Efficient-IoU) and CIoU (Complete-IoU), speeds up the adjustment of the prediction frame and the network regression. In the Neck section of the network, the QAFPN feature fusion extraction network is proposed to replace the PAN-FPN (Path Aggregation Network—Feature Pyramid Network) in YOLOv8s with a Rep-PAN structure based on the QARepNext reparameterization framework for feature fusion extraction to strike a balance between network performance and inference speed. To validate the method, we built a three-axis mobile sorting device and created a dataset of 3000 images of walnuts after shell removal for experiments. The results show that the improved network contains 6071008 parameters, a training time of 2.49 h, a model size of 12.3 MB, an mAP (Mean Average Precision) of 94.5%, and a frame rate of 52.1 FPS. Compared with the original model, the number of parameters decreased by 45.5%, with training time reduced by 32.7%, the model size shrunk by 45.3%, and frame rate improved by 40.8%. However, some accuracy is sacrificed due to the lightweight design, resulting in a 1.2% decrease in mAP. The network reduces the model size by 59.7 MB and 23.9 MB compared to YOLOv7 and YOLOv6, respectively, and improves the frame rate by 15.67 fps and 22.55 fps, respectively. The average confidence and mAP show minimal changes compared to YOLOv7 and improved by 4.2% and 2.4% compared to YOLOv6, respectively. The FastQAFPN-YOLOv8s detection method effectively reduces model size while maintaining recognition accuracy. Full article

In industries spanning manufacturing to software development, defect segmentation is essential for maintaining high standards of product quality and reliability. However, traditional segmentation methods often struggle to accurately identify defects due to challenges like noise interference, occlusion, and feature overlap. To solve these problems, we propose a cross-hierarchy feature fusion network based on a composite dual-channel encoder for surface defect segmentation, called CFF-Net. Specifically, in the encoder of CFF-Net, we design a composite dual-channel module (CDCM), which combines standard convolution with dilated convolution and adopts a dual-path parallel structure to enhance the model’s capability in feature extraction. Then, a dilated residual pyramid module (DRPM) is integrated at the junction of the encoder and decoder, which utilizes the expansion convolution of different expansion rates to effectively capture multi-scale context information. In the final output phase, we introduce a cross-hierarchy feature fusion strategy (CFFS) that combines outputs from different layers or stages, thereby improving the robustness and generalization of the network. Finally, we conducted comparative experiments to evaluate CFF-Net against several mainstream segmentation networks across three distinct datasets: a publicly available Crack500 dataset, a self-built Bearing dataset, and another publicly available SD-saliency-900 dataset. The results demonstrated that CFF-Net consistently outperformed competing methods in segmentation tasks. Specifically, in the Crack500 dataset, CFF-Net achieved notable performance metrics, including an Mcc of 73.36%, Dice coefficient of 74.34%, and Jaccard index of 59.53%. For the Bearing dataset, it recorded an Mcc of 76.97%, Dice coefficient of 77.04%, and Jaccard index of 63.28%. Similarly, in the SD-saliency-900 dataset, CFF-Net achieved an Mcc of 84.08%, Dice coefficient of 85.82%, and Jaccard index of 75.67%. These results underscore CFF-Net’s effectiveness and reliability in handling diverse segmentation challenges across different datasets. Full article

In the field of smart grid monitoring, real-time defect detection for overhead transmission lines is crucial for ensuring the safety and stability of power systems. This paper proposes a defect detection model for overhead transmission lines based on an improved YOLOv8n model, named ABG-YOLOv8n. The model incorporates four key improvements: Lightweight convolutional neural networks and spatial–channel reconstructed convolutional modules are integrated into the backbone network and feature fusion network, respectively. A bidirectional feature pyramid network is employed to achieve multi-scale feature fusion, and the ASFF mechanism is used to enhance the sensitivity of YOLOv8n’s detection head. Finally, comprehensive comparative experiments were conducted with multiple models to validate the effectiveness of the proposed method based on the obtained prediction curves and various performance metrics. The validation results indicate that the proposed ABG-YOLOv8n model achieves a 4.5% improvement in mean average precision compared to the original YOLOv8n model, with corresponding increases of 3.6% in accuracy and 2.0% in recall. Additionally, the ABG-YOLOv8n model demonstrates superior detection performance compared to other enhanced YOLO models. Full article

Sheep facial expressions are valuable indicators of their pain levels, playing a critical role in monitoring their health and welfare. In response to challenges such as missed detections, false positives, and low recognition accuracy in sheep facial expression recognition, this paper introduces an enhanced algorithm based on YOLOv8n, referred to as SimAM-MobileViTAttention-EfficiCIoU-AA2_SPPF-YOLOv8n (SMEA-YOLOv8n). Firstly, the proposed method integrates the parameter-free Similarity-Aware Attention Mechanism (SimAM) and MobileViTAttention modules into the CSP Bottleneck with 2 Convolutions(C2f) module of the neck network, aiming to enhance the model’s feature representation and fusion capabilities in complex environments while mitigating the interference of irrelevant background features. Additionally, the EfficiCIoU loss function replaces the original Complete IoU(CIoU) loss function, thereby improving bounding box localization accuracy and accelerating model convergence. Furthermore, the Spatial Pyramid Pooling-Fast (SPPF) module in the backbone network is refined with the addition of two global average pooling layers, strengthening the extraction of sheep facial expression features and bolstering the model’s core feature fusion capacity. Experimental results reveal that the proposed method achieves a [email protected] of 92.5%, a Recall of 91%, a Precision of 86%, and an F1-score of 88.0%, reflecting improvements of 4.5%, 9.1%, 2.8%, and 6.0%, respectively, compared to the baseline model. Notably, the [email protected] for normal and abnormal sheep facial expressions increased by 3.7% and 5.3%, respectively, demonstrating the method’s effectiveness in enhancing recognition accuracy under complex environmental conditions. Full article

An unescapable trend of hyperspectral image (HSI) has been toward classification with high accuracy and splendid performance. In recent years, Transformers have made remarkable progress in the HSI classification task. However, Transformer-based methods still encounter two main challenges. First, they concentrate on extracting spectral information and are incapable of using spatial information to a great extent. Second, they lack the utilization of multiscale features and do not sufficiently combine the advantages of the Transformer’s global feature extraction and multiscale feature extraction. To tackle these challenges, this article proposes a new solution named the hybrid-scale dual-domain Transformer with crisscrossed interaction (HSD²Former) for HSI classification. HSD²Former consists of three functional modules: dual-dimension multiscale convolutional embedding (D²MSCE), mixed domainFormer (MDFormer), and pyramid scale fusion block (PSFB). D²MSCE supersedes conventional patch embedding to generate spectral and spatial tokens at different scales, effectively enriching the diversity of spectral-spatial features. MDFormer is designed to facilitate self-enhancement and information interaction between the spectral domain and spatial domain, alleviating the heterogeneity of the spatial domain and spectral domain. PSFB introduces a straightforward fusion manner to achieve advanced semantic information for classification. Extensive experiments conducted on four datasets demonstrate the robustness and significance of HSD²Former. The classification evaluation indicators of OA, AA, and Kappa on four datasets almost exceed 98%, reaching state-of-the-art performance. Full article