Search Results (14)

In smart logistics and intelligent manufacturing scenarios, the deployment of Autonomous Guided Vehicles (AGVs) necessitates vision systems that balance stringent real-time constraints with high detection accuracy. However, contemporary lightweight models often struggle with multi-scale feature representation and precision degradation. To address these challenges, this study presents LSOD-YOLO, a tailored evolution of YOLO11n designed for embedded AGV systems. Our methodology focuses on three architectural innovations: (1) we propose a Lightweight Shared Convolution Detection (LSCD) head integrated with Group Normalization (GN) and a scale-adaptive mechanism to harmonize multi-scale feature responses; (2) we re-engineer the backbone using a Star-Net architecture enhanced by Gated MLPs and Depthwise Attention to refine local spatial modeling; and (3) we integrate multi-branch residuals and Channel Attention (CAA) into the C3k2-Star-CAA module to enhance robustness against occlusions and complex backgrounds. The experimental validation on a self-built AGV industrial dataset and COCO128 reveals a compelling performance leap: a 30 FPS increase in throughput and a 1.5% gain in precision, all achieved with 32.8% fewer parameters. These findings confirm that LSOD-YOLO achieves a superior trade-off between computational efficiency and reliability, showing great potential for seamless deployment in resource-constrained AGV visual tasks. Full article

Reliable vision-based sensing of protective equipment wearing compliance is essential for safety monitoring in underground mining environments, where complex lighting conditions, similar background textures, and large variations in the scale of wearable items significantly degrade detection performance. To address these challenges, this study proposes a vision-based protective equipment wearing compliance detection method for underground personnel based on CGALS-YOLO. Traditional object detection models often introduce substantial redundant background information during multi-scale feature fusion, which weakens the perception of key wearing regions, particularly for small-scale targets. To alleviate this issue, a content-guided feature fusion (CGAFusion) module is incorporated into the neck of the YOLOv8 network, enabling adaptive fusion of same-scale multi-path features through the collaborative effects of channel, spatial, and pixel attention mechanisms. This design enhances target-related feature representation while suppressing background interference in complex underground scenes. Furthermore, to reduce parameter redundancy and improve cross-scale discrimination consistency in the detection head, a lightweight shared convolution detection (LSCD) structure is introduced. By employing cross-scale shared convolution parameters, group normalization, and scale-adaptive regression, the proposed model achieves a parameter reduction of approximately 23.9% while lowering computational complexity and maintaining stable multi-scale detection performance. Experimental results on an underground protective equipment wearing compliance dataset demonstrate that CGALS-YOLO improves detection accuracy by approximately 4.6% and recall by 3.1% compared with the baseline YOLOv8n, achieving an mAP@0.5 of 89.4%. These results validate the effectiveness and practical applicability of the proposed method for real-time vision-based safety monitoring in underground environments. Full article

The ocean archives massive, stable remote sensing datasets, and leveraging these data to achieve intelligent real-time recognition of marine organisms has become a core task in the field of marine remote sensing. However, in complex seabed environments, marine monitoring equipment is often constrained by limited computing power—this creates an urgent demand among oceanographers for detection algorithms with low computational complexity, which can be widely deployed on low-cost, simple marine remote sensing devices. To address this demand, this study proposes a deep learning-based algorithm for lightweight seabed organism detection efficiently (LSOD). This algorithm integrates Mamba and YOLO principles to enable efficient lightweight benthic organism detection. For LSOD’s neck, the original concatenation modules are improved, which efficiently aggregates feature layer information across backbone stages for cross-scale fusion. To further reduce the computational requirements of LSOD, a new detection head module based on group normalization and shared convolution operations is designed. These improvements maintain a reasonable computational load while enhancing the precision of the object detection network. EUDD tests indicate LSOD’s performance: the detection precision achieves 90.6% (sea cucumbers), 91.6% (sea urchins), and 93.5% (scallops). Comparisons with mainstream models confirm its superiority in detecting benthic organisms. This work is expected to provide new insights and approaches for intelligent remote sensing and analysis in marine ranches. Full article

Marine oil spill detection using Synthetic Aperture Radar (SAR) is crucial but challenged by dynamic marine conditions, diverse spill scales, and limitations in existing algorithms regarding model size and real-time performance. To address these challenges, we propose LSFE-YOLO, a YOLOv8s-optimized (You Only Look Once version 8) lightweight model with an original, domain-tailored synergistic integration of FasterNet, GN-LSC Head (GroupNorm Lightweight Shared Convolution Head), and C2f_MBE (C2f Mobile Bottleneck Enhanced). FasterNet serves as the backbone (25% neck width reduction), leveraging partial convolution (PConv) to minimize memory access and redundant computations—overcoming traditional lightweight backbones’ high memory overhead—laying the foundation for real-time deployment while preserving feature extraction. The proposed GN-LSC Head replaces YOLOv8’s decoupled head: its shared convolutions reduce parameter redundancy by approximately 40%, and GroupNorm (Group Normalization) ensures stable accuracy under edge computing’s small-batch constraints, outperforming BatchNorm (Batch Normalization) in resource-limited scenarios. The C2f_MBE module integrates EffectiveSE (Effective Squeeze and Excitation)-optimized MBConv (Mobile Inverted Bottleneck Convolution) into C2f: MBConv’s inverted-residual design enhances multi-scale feature capture, while lightweight EffectiveSE strengthens discriminative oil spill features without extra computation, addressing the original C2f’s scale variability insufficiency. Additionally, an SE (Squeeze and Excitation) attention mechanism embedded upstream of SPPF (Spatial Pyramid Pooling Fast) suppresses background interference (e.g., waves, biological oil films), synergizing with FasterNet and C2f_MBE to form a cascaded feature optimization pipeline that refines representations throughout the model. Experimental results show that LSFE-YOLO improves mAP (mean Average Precision) by 1.3% and F1 score by 1.7% over YOLOv8s, while achieving substantial reductions in model size (81.9%), parameter count (82.9%), and computational cost (84.2%), alongside a 20 FPS (Frames Per Second) increase in detection speed. LSFE-YOLO offers an efficient and effective solution for real-time marine oil spill detection. Full article

The rapid development of intelligent transportation systems and autonomous driving technologies has made visual perception a key component in ensuring safety and improving efficiency in complex traffic environments. As a core task in visual perception, object detection directly affects the reliability of downstream modules such as path planning and decision control. However, adverse weather conditions (e.g., fog, rain, and snow) significantly degrade image quality—causing texture blurring, reduced contrast, and increased noise—which in turn weakens the robustness of traditional detection models and raises potential traffic safety risks. To address this challenge, this paper proposes an enhanced object detection framework, ECL-YOLOv11 (Edge-enhanced, Context-guided, and Lightweight YOLOv11), designed to improve detection accuracy and real-time performance under adverse weather conditions, thereby providing a reliable solution for in-vehicle perception systems. The ECL-YOLOv11 architecture integrates three key modules: (1) a Convolutional Edge-enhancement (CE) module that fuses edge features extracted by Sobel operators with convolutional features to explicitly retain boundary and contour information, thereby alleviating feature degradation and improving localization accuracy under low-visibility conditions; (2) a Context-guided Multi-scale Fusion Network (AENet) that enhances perception of small and distant objects through multi-scale feature integration and context modeling, improving semantic consistency and detection stability in complex scenes; and (3) a Lightweight Shared Convolutional Detection Head (LDHead) that adopts shared convolutions and GroupNorm normalization to optimize computational efficiency, reduce inference latency, and satisfy the real-time requirements of on-board systems. Experimental results show that ECL-YOLOv11 achieves mAP@50 and mAP@50–95 values of 62.7% and 40.5%, respectively, representing improvements of 1.3% and 0.8% over the baseline YOLOv11, while the Precision reaches 73.1%. The model achieves a balanced trade-off between accuracy and inference speed, operating at 237.8 FPS on standard hardware. Ablation studies confirm the independent effectiveness of each proposed module in feature enhancement, multi-scale fusion, and lightweight detection, while their integration further improves overall performance. Qualitative visualizations demonstrate that ECL-YOLOv11 maintains high-confidence detections across varying motion states and adverse weather conditions, avoiding category confusion and missed detections. These results indicate that the proposed framework provides a reliable and adaptable foundation for all-weather perception in autonomous driving systems, ensuring both operational safety and real-time responsiveness. Full article

The acquisition of maize tassel phenotypic information is important for studying maize growth and improving yield. Unfortunately, tassel detection remains challenging due to complex field conditions, including image resolution, light variation, cultivar differences, and planting density. Existing models still have limitations in detecting small, overlapping, and blurred tassels in such environments, and their size and complexity restrict deployment on mobile devices. To address this, a lightweight model, RALSD-YOLO, is developed based on the YOLOv8n architecture. In this study, a lightweight module, C2f_RVB_EMA, is introduced to replace C2f in the backbone network, reducing shallow-layer noise and lowering the number of parameters. Adown and SPPF_LSKA are used to replace Conv and SPPF, respectively, mitigating feature loss during downsampling and enhancing feature extraction in complex environments. The neck incorporates GSConv combined with the VoVGSCSP module to form a streamlined structure, achieving model compression and a lightweight design. Finally, a lightweight shared convolution detection head, LiSCDetect, is developed with group normalization, shared convolutions, and DFL post-processing, enabling feature sharing across layers and improving the precision and efficiency of small object detection. To evaluate model performance, RALSD-YOLO was compared with Faster R-CNN, SSD, YOLOv3, YOLOv5n, YOLOv7-tiny, and YOLOv8 under the same conditions. RALSD-YOLO achieved a precision of 96.8%, a recall of 97.9%, an mAP50 of 98.9%, and an F1 score of 97.3%, representing improvements of 1.3%, 2.1%, 0.4%, and 1.7%, respectively, over YOLOv8n while reducing parameter count and computational cost to 59.47% and 54.32% of those of YOLOv8n. Furthermore, on the MTDC dataset, RALSD-YOLO achieved a precision of 92.7%, a recall of 87.9%, an mAP50 of 92.1%, and an F1 score of 90.24%. The model combines high precision with a small size, making it suitable for lightweight deployment and providing effective support for intelligent maize management and harvesting decisions. Full article

Underwater debris detection plays a crucial role in marine environmental protection. However, existing object detection algorithms generally suffer from excessive model complexity and insufficient detection accuracy, making it difficult to meet the real-time detection requirements in resource-constrained underwater environments. To address this challenge, this paper proposes a novel lightweight object detection network named the Shared Pyramid Lightweight Network (SPyramidLightNet). The network adopts an improved architecture based on YOLOv11 and achieves an optimal balance between detection performance and computational efficiency by integrating three core innovative modules. First, the Split–Merge Attention Block (SMAB) employs a dynamic kernel selection mechanism and split–merge strategy, significantly enhancing feature representation capability through adaptive multi-scale feature fusion. Second, the C3 GroupNorm Detection Head (C3GNHead) introduces a shared convolution mechanism and GroupNorm normalization strategy, substantially reducing the computational complexity of the detection head while maintaining detection accuracy. Finally, the Shared Pyramid Convolution (SPyramidConv) replaces traditional pooling operations with a parameter-sharing multi-dilation-rate convolution architecture, achieving more refined and efficient multi-scale feature aggregation. Extensive experiments on underwater debris datasets demonstrate that SPyramidLightNet achieves 0.416 on the mAP@0.5:0.95 metric, significantly outperforming mainstream algorithms including Faster-RCNN, SSD, RT-DETR, and the YOLO series. Meanwhile, compared to the baseline YOLOv11, the proposed algorithm achieves an 11.8% parameter compression and a 17.5% computational complexity reduction, with an inference speed reaching 384 FPS, meeting the stringent requirements for real-time detection. Ablation experiments and visualization analyses further validate the effectiveness and synergistic effects of each core module. This research provides important theoretical guidance for the design of lightweight object detection algorithms and lays a solid foundation for the development of automated underwater debris recognition and removal technologies. Full article

Transmission lines are widely distributed in complex environments, making them susceptible to foreign object intrusion, which could lead to serious consequences, i.e., power outages. Currently, foreign object detection on transmission lines is primarily conducted through UAV-based field inspections. However, the captured data must be transmitted back to a central facility for analysis, resulting in low efficiency and the inability to perform real-time, industrial-grade detection. Although recent YOLO series models can be deployed on UAVs for object detection, these models’ substantial computational requirements often exceed the processing capabilities of UAV platforms, limiting their ability to perform real-time inference tasks. In this study, we propose a novel lightweight detection algorithm, SCL-YOLOv8, which is based on the original YOLO model. We introduce StarNet to replace the CSPDarknet53 backbone as the feature extraction network, thereby reducing computational complexity while maintaining high feature extraction efficiency. We design a lightweight module, CGLU-ConvFormer, which enhances multi-scale feature representation and local feature extraction by integrating convolutional operations with gating mechanisms. Furthermore, the detection head of the original YOLO model is improved by introducing shared convolutional layers and group normalization, which helps reduce redundant computations and enhances multi-scale feature fusion. Experimental results demonstrate that the proposed model not only improves the detection accuracy but also significantly reduces the number of model parameters. Specifically, SCL-YOLOv8 achieves a mAP@0.5 of 94.2% while reducing the number of parameters by 56.8%, FLOPS by 45.7%, and model size by 50% compared with YOLOv8n. Full article

Accurate real-time detection of hawthorn by vision systems is a fundamental prerequisite for automated harvesting. This study addresses the challenges in hawthorn orchards—including target overlap, leaf occlusion, and environmental variations—which lead to compromised detection accuracy, high computational resource demands, and poor real-time performance in existing methods. To overcome these limitations, we propose YOLO-DCL (group shuffling convolution and coordinate attention integrated with a lightweight head based on YOLOv8n), a novel lightweight hawthorn detection model. The backbone network employs dynamic group shuffling convolution (DGCST) for efficient and effective feature extraction. Within the neck network, coordinate attention (CA) is integrated into the feature pyramid network (FPN), forming an enhanced multi-scale feature pyramid network (HSPFN); this integration further optimizes the C2f structure. The detection head is designed utilizing shared convolution and batch normalization to streamline computation. Additionally, the PIoUv2 (powerful intersection over union version 2) loss function is introduced to significantly reduce model complexity. Experimental validation demonstrates that YOLO-DCL achieves a precision of 91.6%, recall of 90.1%, and mean average precision (mAP) of 95.6%, while simultaneously reducing the model size to 2.46 MB with only 1.2 million parameters and 4.8 GFLOPs computational cost. To rigorously assess real-world applicability, we developed and deployed a detection system based on the PySide6 framework on an NVIDIA Jetson Xavier NX edge device. Field testing validated the model’s robustness, high accuracy, and real-time performance, confirming its suitability for integration into harvesting robots operating in practical orchard environments. Full article

Accurate and real-time detection of rice shelling rate (SR) and brown rice breakage rate (BR) is crucial for intelligent hulling sorting but remains challenging because of small grain size, dense adhesion, and uneven illumination causing missed detections and blurred boundaries in traditional YOLOv8n. This paper proposes a high-precision, lightweight solution based on an enhanced YOLOv8n with improvements in network architecture, feature fusion, and attention mechanism. The backbone’s C2f module is replaced with C2f-Faster-CGLU, integrating partial convolution (PConv) local convolution and convolutional gated linear unit (CGLU) gating to reduce computational redundancy via sparse interaction and enhance small-target feature extraction. A bidirectional feature pyramid network (BiFPN) weights multiscale feature fusion to improve edge positioning accuracy of dense grains. Attention mechanism for fine-grained classification (AFGC) is embedded to focus on texture and damage details, enhancing adaptability to light fluctuations. The Detect_Rice lightweight head compresses parameters via group normalization and dynamic convolution sharing, optimizing small-target response. The improved model achieved 96.8% precision and 96.2% mAP. Combined with a quantity–mass model, SR/BR detection errors reduced to 1.11% and 1.24%, meeting national standard (GB/T 29898-2013) requirements, providing an effective real-time solution for intelligent hulling sorting. Full article

To improve the performance of vehicle target detection in complex traffic environments and solve the problem that it is difficult to make a lightweight detection model, this paper proposes a lightweight vehicle detection model based on enhanced You Only Look Once v8. This method improves the feature extraction aggregation network by introducing an Adaptive Downsampling module, which can dynamically adjust the downsampling method, thereby increasing the model’s attention to key features, especially for small objects and occluded objects, while maintaining a lightweight structure, effectively reducing the model complexity while improving detection accuracy. A Lightweight Shared Convolution Detection Head was designed. By designing a shared convolution layer through group normalization, the detection head of the original model was improved, which can reduce redundant calculations and parameters and enhance the ability of global information fusion between feature maps, thereby achieving the purpose of improving computational efficiency. When tested in the KITTI 2D and UA-DETRAC datasets, the mAP of the proposed model was improved by 1.1% and 2.0%, respectively, the FPS was improved by 12% and 11%, respectively, the number of parameters was reduced by 33%, and the FLOPs were reduced by 28%. Full article

Clubbing finger is a significant clinical indicator, and its early detection is essential for the diagnosis and treatment of associated diseases. However, traditional diagnostic methods rely heavily on the clinician’s subjective assessment, which can be prone to biases and may lack standardized tools. Unlike other diagnostic challenges, the characteristic changes of clubbing finger are subtle and localized, necessitating high-precision feature extraction. Existing models often fail to capture these delicate changes accurately, potentially missing crucial diagnostic features or generating false positives. Furthermore, these models are often not suited for accurate clinical diagnosis in resource-constrained settings. To address these challenges, we propose MSG-YOLO, a lightweight clubbing finger detection model based on YOLOv8n, designed to enhance both detection accuracy and efficiency. The model first employs a multi-scale dilated residual module, which expands the receptive field using dilated convolutions and residual connections, thereby improving the model’s ability to capture features across various scales. Additionally, we introduce a Selective Feature Fusion Pyramid Network (SFFPN) that dynamically selects and enhances critical features, optimizing the flow of information while minimizing redundancy. To further refine the architecture, we reconstruct the YOLOv8 detection head with group normalization and shared-parameter convolutions, significantly reducing the model’s parameter count and increasing computational efficiency. Experimental results indicate that the model maintains high detection accuracy with reduced parameter and computational requirements. Compared to YOLOv8n, MSG-YOLO achieves a 48.74% reduction in parameter count and a 24.17% reduction in computational load, while improving the mAP0.5 score by 2.86%, reaching 93.64%. This algorithm strikes a balance between accuracy and lightweight design, offering efficient and reliable clubbing finger detection even in resource-constrained environments. Full article

To address the issues of target feature blurring and increased false detections caused by high compression rates in deepfake videos, as well as the high computational resource requirements of existing face extractors, we propose a lightweight face extractor to assist deepfake detection, GCS-YOLOv8. Firstly, we employ the HGStem module for initial downsampling to address the issue of false detections of small non-face objects in deepfake videos, thereby improving detection accuracy. Secondly, we introduce the C2f-GDConv module to mitigate the low-FLOPs pitfall while reducing the model’s parameters, thereby lightening the network. Additionally, we add a new P6 large target detection layer to expand the receptive field and capture multi-scale features, solving the problem of detecting large-scale faces in low-compression deepfake videos. We also design a cross-scale feature fusion module called CCFG (CNN-based Cross-Scale Feature Fusion with GDConv), which integrates features from different scales to enhance the model’s adaptability to scale variations while reducing network parameters, addressing the high computational resource requirements of traditional face extractors. Furthermore, we improve the detection head by utilizing group normalization and shared convolution, simplifying the process of face detection while maintaining detection performance. The training dataset was also refined by removing low-accuracy and low-resolution labels, which reduced the false detection rate. Experimental results demonstrate that, compared to YOLOv8, this face extractor achieves the AP of 0.942, 0.927, and 0.812 on the WiderFace dataset’s Easy, Medium, and Hard subsets, representing improvements of 1.1%, 1.3%, and 3.7% respectively. The model’s parameters and FLOPs are only 1.68 MB and 3.5 G, reflecting reductions of 44.2% and 56.8%, making it more effective and lightweight in extracting faces from deepfake videos. Full article

In order to detect rice false smut quickly and accurately, a lightweight false smut detection model, YOLOv8n-MBS, was proposed in this study. The model introduces the C2f_MSEC module to replace C2f in the backbone network for better extraction of key features of false smut, enhances the feature fusion capability of the neck network for different sizes of false smut by using a weighted bidirectional feature pyramid network, and designs a group-normalized shared convolution lightweight detection head to reduce the number of parameters in the model head to achieve model lightweight. The experimental results show that YOLOv8n-MBS has an average accuracy of 93.9%, a parameter count of 1.4 M, and a model size of 3.3 MB. Compared with the SSD model, the average accuracy of the model in this study increased by 4%, the number of parameters decreased by 89.8%, and the model size decreased by 86.9%; compared with the YOLO series of YOLOv7-tiny, YOLOv5n, YOLOv5s, and YOLOv8n models, the YOLOv8n-MBS model showed outstanding performance in terms of model accuracy and model performance detection; compared to the latest YOLOv9t and YOLOv10n models, the average model accuracy increased by 2.8% and 2.2%, the number of model parameters decreased by 30% and 39.1%, and the model size decreased by 29.8% and 43.1%, respectively. This method enables more accurate and lighter-weight detection of false smut, which provides the basis for intelligent management of rice blast disease in the field and thus promotes food security. Full article