Human object detection and recognition is essential for elderly monitoring and assisted living however,models relying solely on pose or scene context often struggle in cluttered or visually ambiguous settings.To addre...Human object detection and recognition is essential for elderly monitoring and assisted living however,models relying solely on pose or scene context often struggle in cluttered or visually ambiguous settings.To address this,we present SCENET-3D,a transformer-drivenmultimodal framework that unifies human-centric skeleton features with scene-object semantics for intelligent robotic vision through a three-stage pipeline.In the first stage,scene analysis,rich geometric and texture descriptors are extracted from RGB frames,including surface-normal histograms,angles between neighboring normals,Zernike moments,directional standard deviation,and Gabor-filter responses.In the second stage,scene-object analysis,non-human objects are segmented and represented using local feature descriptors and complementary surface-normal information.In the third stage,human-pose estimation,silhouettes are processed through an enhanced MoveNet to obtain 2D anatomical keypoints,which are fused with depth information and converted into RGB-based point clouds to construct pseudo-3D skeletons.Features from all three stages are fused and fed in a transformer encoder with multi-head attention to resolve visually similar activities.Experiments on UCLA(95.8%),ETRI-Activity3D(89.4%),andCAD-120(91.2%)demonstrate that combining pseudo-3D skeletonswith rich scene-object fusion significantly improves generalizable activity recognition,enabling safer elderly care,natural human–robot interaction,and robust context-aware robotic perception in real-world environments.展开更多
Visible and infrared(RGB-IR)fusion object detection plays an important role in security,disaster relief,etc.In recent years,deep-learning-based RGB-IR fusion detection methods have been developing rapidly,but still st...Visible and infrared(RGB-IR)fusion object detection plays an important role in security,disaster relief,etc.In recent years,deep-learning-based RGB-IR fusion detection methods have been developing rapidly,but still struggle to deal with the complex and changing scenarios captured by drones,mainly due to two reasons:(A)RGB-IR fusion detectors are susceptible to inferior inputs that degrade performance and stability.(B)RGB-IR fusion detectors are susceptible to redundant features that reduce accuracy and efficiency.In this paper,an innovative RGB-IR fusion detection framework based on global-local feature optimization,named GLFDet,is proposed to improve the detection performance and efficiency of drone-captured objects.The key components of GLFDet include a Global Feature Optimization(GFO)module,a Local Feature Optimization(LFO)module and a Channel Separation Fusion(CSF)module.Specifically,GFO calculates the information content of the input image from the frequency domain and optimizes the features holistically.Then,LFO dynamically selects high-value features and filters out low-value features before fusion,which significantly improves the efficiency of fusion.Finally,CSF fuses the RGB and IR features across the corresponding channels,which avoids the rearrangement of the channel relationships and enhances the model stability.Extensive experimental results show that the proposed method achieves the best performance on three popular RGB-IR datasets Drone Vehicle,VEDAI,and LLVIP.In addition,GLFDet is more lightweight than other comparable models,making it more appealing to edge devices such as drones.The code is available at https://github.com/lao chen330/GLFDet.展开更多
Dear Editor,This letter focuses on the fact that small objects with few pixels disappear in feature maps with large receptive fields, as the network deepens, in object detection tasks. Therefore, the detection of dens...Dear Editor,This letter focuses on the fact that small objects with few pixels disappear in feature maps with large receptive fields, as the network deepens, in object detection tasks. Therefore, the detection of dense small objects is challenging.展开更多
Deep learning has made significant progress in the field of oriented object detection for remote sensing images.However,existing methods still face challenges when dealing with difficult tasks such as multi-scale targ...Deep learning has made significant progress in the field of oriented object detection for remote sensing images.However,existing methods still face challenges when dealing with difficult tasks such as multi-scale targets,complex backgrounds,and small objects in remote sensing.Maintaining model lightweight to address resource constraints in remote sensing scenarios while improving task completion for remote sensing tasks remains a research hotspot.Therefore,we propose an enhanced multi-scale feature extraction lightweight network EM-YOLO based on the YOLOv8s architecture,specifically optimized for the characteristics of large target scale variations,diverse orientations,and numerous small objects in remote sensing images.Our innovations lie in two main aspects:First,a dynamic snake convolution(DSC)is introduced into the backbone network to enhance the model’s feature extraction capability for oriented targets.Second,an innovative focusing-diffusion module is designed in the feature fusion neck to effectively integrate multi-scale feature information.Finally,we introduce Layer-Adaptive Sparsity for magnitude-based Pruning(LASP)method to perform lightweight network pruning to better complete tasks in resource-constrained scenarios.Experimental results on the lightweight platform Orin demonstrate that the proposed method significantly outperforms the original YOLOv8s model in oriented remote sensing object detection tasks,and achieves comparable or superior performance to state-of-the-art methods on three authoritative remote sensing datasets(DOTA v1.0,DOTA v1.5,and HRSC2016).展开更多
In modern industrial production,foreign object detection in complex environments is crucial to ensure product quality and production safety.Detection systems based on deep-learning image processing algorithms often fa...In modern industrial production,foreign object detection in complex environments is crucial to ensure product quality and production safety.Detection systems based on deep-learning image processing algorithms often face challenges with handling high-resolution images and achieving accurate detection against complex backgrounds.To address these issues,this study employs the PatchCore unsupervised anomaly detection algorithm combined with data augmentation techniques to enhance the system’s generalization capability across varying lighting conditions,viewing angles,and object scales.The proposed method is evaluated in a complex industrial detection scenario involving the bogie of an electric multiple unit(EMU).A dataset consisting of complex backgrounds,diverse lighting conditions,and multiple viewing angles is constructed to validate the performance of the detection system in real industrial environments.Experimental results show that the proposed model achieves an average area under the receiver operating characteristic curve(AUROC)of 0.92 and an average F1 score of 0.85.Combined with data augmentation,the proposed model exhibits improvements in AUROC by 0.06 and F1 score by 0.03,demonstrating enhanced accuracy and robustness for foreign object detection in complex industrial settings.In addition,the effects of key factors on detection performance are systematically analyzed,providing practical guidance for parameter selection in real industrial applications.展开更多
Small object detection has been a focus of attention since the emergence of deep learning-based object detection.Although classical object detection frameworks have made significant contributions to the development of...Small object detection has been a focus of attention since the emergence of deep learning-based object detection.Although classical object detection frameworks have made significant contributions to the development of object detection,there are still many issues to be resolved in detecting small objects due to the inherent complexity and diversity of real-world visual scenes.In particular,the YOLO(You Only Look Once)series of detection models,renowned for their real-time performance,have undergone numerous adaptations aimed at improving the detection of small targets.In this survey,we summarize the state-of-the-art YOLO-based small object detection methods.This review presents a systematic categorization of YOLO-based approaches for small-object detection,organized into four methodological avenues,namely attention-based feature enhancement,detection-head optimization,loss function,and multi-scale feature fusion strategies.We then examine the principal challenges addressed by each category.Finally,we analyze the performance of thesemethods on public benchmarks and,by comparing current approaches,identify limitations and outline directions for future research.展开更多
In recent years,with the rapid advancement of artificial intelligence,object detection algorithms have made significant strides in accuracy and computational efficiency.Notably,research and applications of Anchor-Free...In recent years,with the rapid advancement of artificial intelligence,object detection algorithms have made significant strides in accuracy and computational efficiency.Notably,research and applications of Anchor-Free models have opened new avenues for real-time target detection in optical remote sensing images(ORSIs).However,in the realmof adversarial attacks,developing adversarial techniques tailored to Anchor-Freemodels remains challenging.Adversarial examples generated based on Anchor-Based models often exhibit poor transferability to these new model architectures.Furthermore,the growing diversity of Anchor-Free models poses additional hurdles to achieving robust transferability of adversarial attacks.This study presents an improved cross-conv-block feature fusion You Only Look Once(YOLO)architecture,meticulously engineered to facilitate the extraction ofmore comprehensive semantic features during the backpropagation process.To address the asymmetry between densely distributed objects in ORSIs and the corresponding detector outputs,a novel dense bounding box attack strategy is proposed.This approach leverages dense target bounding boxes loss in the calculation of adversarial loss functions.Furthermore,by integrating translation-invariant(TI)and momentum-iteration(MI)adversarial methodologies,the proposed framework significantly improves the transferability of adversarial attacks.Experimental results demonstrate that our method achieves superior adversarial attack performance,with adversarial transferability rates(ATR)of 67.53%on the NWPU VHR-10 dataset and 90.71%on the HRSC2016 dataset.Compared to ensemble adversarial attack and cascaded adversarial attack approaches,our method generates adversarial examples in an average of 0.64 s,representing an approximately 14.5%improvement in efficiency under equivalent conditions.展开更多
Recognising human-object interactions(HOI)is a challenging task for traditional machine learning models,including convolutional neural networks(CNNs).Existing models show limited transferability across complex dataset...Recognising human-object interactions(HOI)is a challenging task for traditional machine learning models,including convolutional neural networks(CNNs).Existing models show limited transferability across complex datasets such as D3D-HOI and SYSU 3D HOI.The conventional architecture of CNNs restricts their ability to handle HOI scenarios with high complexity.HOI recognition requires improved feature extraction methods to overcome the current limitations in accuracy and scalability.This work proposes a Novel quantum gate-enabled hybrid CNN(QEH-CNN)for effectiveHOI recognition.Themodel enhancesCNNperformance by integrating quantumcomputing components.The framework begins with bilateral image filtering,followed bymulti-object tracking(MOT)and Felzenszwalb superpixel segmentation.A watershed algorithm refines object boundaries by cleaning merged superpixels.Feature extraction combines a histogram of oriented gradients(HOG),Global Image Statistics for Texture(GIST)descriptors,and a novel 23-joint keypoint extractionmethod using relative joint angles and joint proximitymeasures.A fuzzy optimization process refines the extracted features before feeding them into the QEH-CNNmodel.The proposed model achieves 95.06%accuracy on the 3D-D3D-HOI dataset and 97.29%on the SYSU3DHOI dataset.Theintegration of quantum computing enhances feature optimization,leading to improved accuracy and overall model efficiency.展开更多
Salient object detection(SOD)models struggle to simultaneously preserve global structure,maintain sharp object boundaries,and sustain computational efficiency in complex scenes.In this study,we propose SPSALNet,a task...Salient object detection(SOD)models struggle to simultaneously preserve global structure,maintain sharp object boundaries,and sustain computational efficiency in complex scenes.In this study,we propose SPSALNet,a task-driven two-stage(macro–micro)architecture that restructures the SOD process around superpixel representations.In the proposed approach,a“split-and-enhance”principle,introduced to our knowledge for the first time in the SOD literature,hierarchically classifies superpixels and then applies targeted refinement only to ambiguous or error-prone regions.At the macro stage,the image is partitioned into content-adaptive superpixel regions,and each superpixel is represented by a high-dimensional region-level feature vector.These representations define a regional decomposition problem in which superpixels are assigned to three classes:background,object interior,and transition regions.Superpixel tokens interact with a global feature vector from a deep network backbone through a cross-attention module and are projected into an enriched embedding space that jointly encodes local topology and global context.At the micro stage,the model employs a U-Net-based refinement process that allocates computational resources only to ambiguous transition regions.The image and distance–similarity maps derived from superpixels are processed through a dual-encoder pathway.Subsequently,channel-aware fusion blocks adaptively combine information from these two sources,producing sharper and more stable object boundaries.Experimental results show that SPSALNet achieves high accuracy with lower computational cost compared to recent competing methods.On the PASCAL-S and DUT-OMRON datasets,SPSALNet exhibits a clear performance advantage across all key metrics,and it ranks first on accuracy-oriented measures on HKU-IS.On the challenging DUT-OMRON benchmark,SPSALNet reaches a MAE of 0.034.Across all datasets,it preserves object boundaries and regional structure in a stable and competitive manner.展开更多
The ubiquity of mobile devices has driven advancements in mobile object detection.However,challenges in multi-scale object detection in open,complex environments persist due to limited computational resources.Traditio...The ubiquity of mobile devices has driven advancements in mobile object detection.However,challenges in multi-scale object detection in open,complex environments persist due to limited computational resources.Traditional approaches like network compression,quantization,and lightweight design often sacrifice accuracy or feature representation robustness.This article introduces the Fast Multi-scale Channel Shuffling Network(FMCSNet),a novel lightweight detection model optimized for mobile devices.FMCSNet integrates a fully convolutional Multilayer Perceptron(MLP)module,offering global perception without significantly increasing parameters,effectively bridging the gap between CNNs and Vision Transformers.FMCSNet achieves a delicate balance between computation and accuracy mainly by two key modules:the ShiftMLP module,including a shift operation and an MLP module,and a Partial group Convolutional(PGConv)module,reducing computation while enhancing information exchange between channels.With a computational complexity of 1.4G FLOPs and 1.3M parameters,FMCSNet outperforms CNN-based and DWConv-based ShuffleNetv2 by 1%and 4.5%mAP on the Pascal VOC 2007 dataset,respectively.Additionally,FMCSNet achieves a mAP of 30.0(0.5:0.95 IoU threshold)with only 2.5G FLOPs and 2.0M parameters.It achieves 32 FPS on low-performance i5-series CPUs,meeting real-time detection requirements.The versatility of the PGConv module’s adaptability across scenarios further highlights FMCSNet as a promising solution for real-time mobile object detection.展开更多
To address the issues of frequent identity switches(IDs)and degraded identification accuracy in multi object tracking(MOT)under complex occlusion scenarios,this study proposes an occlusion-robust tracking framework ba...To address the issues of frequent identity switches(IDs)and degraded identification accuracy in multi object tracking(MOT)under complex occlusion scenarios,this study proposes an occlusion-robust tracking framework based on face-pedestrian joint feature modeling.By constructing a joint tracking model centered on“intra-class independent tracking+cross-category dynamic binding”,designing a multi-modal matching metric with spatio-temporal and appearance constraints,and innovatively introducing a cross-category feature mutual verification mechanism and a dual matching strategy,this work effectively resolves performance degradation in traditional single-category tracking methods caused by short-term occlusion,cross-camera tracking,and crowded environments.Experiments on the Chokepoint_Face_Pedestrian_Track test set demonstrate that in complex scenes,the proposed method improves Face-Pedestrian Matching F1 area under the curve(F1 AUC)by approximately 4 to 43 percentage points compared to several traditional methods.The joint tracking model achieves overall performance metrics of IDF1:85.1825%and MOTA:86.5956%,representing improvements of 0.91 and 0.06 percentage points,respectively,over the baseline model.Ablation studies confirm the effectiveness of key modules such as the Intersection over Area(IoA)/Intersection over Union(IoU)joint metric and dynamic threshold adjustment,validating the significant role of the cross-category identity matching mechanism in enhancing tracking stability.Our_model shows a 16.7%frame per second(FPS)drop vs.fairness of detection and re-identification in multiple object tracking(FairMOT),with its cross-category binding module adding aboute 10%overhead,yet maintains near-real-time performance for essential face-pedestrian tracking at small resolutions.展开更多
With the rapid expansion of drone applications,accurate detection of objects in aerial imagery has become crucial for intelligent transportation,urban management,and emergency rescue missions.However,existing methods ...With the rapid expansion of drone applications,accurate detection of objects in aerial imagery has become crucial for intelligent transportation,urban management,and emergency rescue missions.However,existing methods face numerous challenges in practical deployment,including scale variation handling,feature degradation,and complex backgrounds.To address these issues,we propose Edge-enhanced and Detail-Capturing You Only Look Once(EHDC-YOLO),a novel framework for object detection in Unmanned Aerial Vehicle(UAV)imagery.Based on the You Only Look Once version 11 nano(YOLOv11n)baseline,EHDC-YOLO systematically introduces several architectural enhancements:(1)a Multi-Scale Edge Enhancement(MSEE)module that leverages multi-scale pooling and edge information to enhance boundary feature extraction;(2)an Enhanced Feature Pyramid Network(EFPN)that integrates P2-level features with Cross Stage Partial(CSP)structures and OmniKernel convolutions for better fine-grained representation;and(3)Dynamic Head(DyHead)with multi-dimensional attention mechanisms for enhanced cross-scale modeling and perspective adaptability.Comprehensive experiments on the Vision meets Drones for Detection(VisDrone-DET)2019 dataset demonstrate that EHDC-YOLO achieves significant improvements,increasing mean Average Precision(mAP)@0.5 from 33.2%to 46.1%(an absolute improvement of 12.9 percentage points)and mAP@0.5:0.95 from 19.5%to 28.0%(an absolute improvement of 8.5 percentage points)compared with the YOLOv11n baseline,while maintaining a reasonable parameter count(2.81 M vs the baseline’s 2.58 M).Further ablation studies confirm the effectiveness of each proposed component,while visualization results highlight EHDC-YOLO’s superior performance in detecting objects and handling occlusions in complex drone scenarios.展开更多
Modern manufacturing processes have become more reliant on automation because of the accelerated transition from Industry 3.0 to Industry 4.0.Manual inspection of products on assembly lines remains inefficient,prone t...Modern manufacturing processes have become more reliant on automation because of the accelerated transition from Industry 3.0 to Industry 4.0.Manual inspection of products on assembly lines remains inefficient,prone to errors and lacks consistency,emphasizing the need for a reliable and automated inspection system.Leveraging both object detection and image segmentation approaches,this research proposes a vision-based solution for the detection of various kinds of tools in the toolkit using deep learning(DL)models.Two Intel RealSense D455f depth cameras were arranged in a top down configuration to capture both RGB and depth images of the toolkits.After applying multiple constraints and enhancing them through preprocessing and augmentation,a dataset consisting of 3300 annotated RGB-D photos was generated.Several DL models were selected through a comprehensive assessment of mean Average Precision(mAP),precision-recall equilibrium,inference latency(target≥30 FPS),and computational burden,resulting in a preference for YOLO and Region-based Convolutional Neural Networks(R-CNN)variants over ViT-based models due to the latter’s increased latency and resource requirements.YOLOV5,YOLOV8,YOLOV11,Faster R-CNN,and Mask R-CNN were trained on the annotated dataset and evaluated using key performance metrics(Recall,Accuracy,F1-score,and Precision).YOLOV11 demonstrated balanced excellence with 93.0%precision,89.9%recall,and a 90.6%F1-score in object detection,as well as 96.9%precision,95.3%recall,and a 96.5%F1-score in instance segmentation with an average inference time of 25 ms per frame(≈40 FPS),demonstrating real-time performance.Leveraging these results,a YOLOV11-based windows application was successfully deployed in a real-time assembly line environment,where it accurately processed live video streams to detect and segment tools within toolkits,demonstrating its practical effectiveness in industrial automation.The application is capable of precisely measuring socket dimensions by utilising edge detection techniques on YOLOv11 segmentation masks,in addition to detection and segmentation.This makes it possible to do specification-level quality control right on the assembly line,which improves the ability to examine things in real time.The implementation is a big step forward for intelligent manufacturing in the Industry 4.0 paradigm.It provides a scalable,efficient,and accurate way to do automated inspection and dimensional verification activities.展开更多
The goal of the present work is to demonstrate the potential of Artificial Neural Network(ANN)-driven Genetic Algorithm(GA)methods for energy efficiency and economic performance optimization of energy efficiency measu...The goal of the present work is to demonstrate the potential of Artificial Neural Network(ANN)-driven Genetic Algorithm(GA)methods for energy efficiency and economic performance optimization of energy efficiency measures in a multi-family house building in Greece.The energy efficiency measures include different heating/cooling systems(such as low-temperature and high-temperature heat pumps,natural gas boilers,split units),building envelope components for floor,walls,roof and windows of variable heat transfer coefficients,the installation of solar thermal collectors and PVs.The calculations of the building loads and investment and operating and maintenance costs of the measures are based on the methodology defined in Directive 2010/31/EU,while economic assumptions are based on EN 15459-1 standard.Typically,multi-objective optimization of energy efficiency measures often requires the simulation of very large numbers of cases involving numerous possible combinations,resulting in intense computational load.The results of the study indicate that ANN-driven GA methods can be used as an alternative,valuable tool for reliably predicting the optimal measures which minimize primary energy consumption and life cycle cost of the building with greatly reduced computational requirements.Through GA methods,the computational time needed for obtaining the optimal solutions is reduced by 96.4%-96.8%.展开更多
Drone-based small object detection is of great significance in practical applications such as military actions, disaster rescue, transportation, etc. However, the severe scale differences in objects captured by drones...Drone-based small object detection is of great significance in practical applications such as military actions, disaster rescue, transportation, etc. However, the severe scale differences in objects captured by drones and lack of detail information for small-scale objects make drone-based small object detection a formidable challenge. To address these issues, we first develop a mathematical model to explore how changing receptive fields impacts the polynomial fitting results. Subsequently, based on the obtained conclusions, we propose a simple but effective Hybrid Receptive Field Network (HRFNet), whose modules include Hybrid Feature Augmentation (HFA), Hybrid Feature Pyramid (HFP) and Dual Scale Head (DSH). Specifically, HFA employs parallel dilated convolution kernels of different sizes to extend shallow features with different receptive fields, committed to improving the multi-scale adaptability of the network;HFP enhances the perception of small objects by capturing contextual information across layers, while DSH reconstructs the original prediction head utilizing a set of high-resolution features and ultrahigh-resolution features. In addition, in order to train HRFNet, the corresponding dual-scale loss function is designed. Finally, comprehensive evaluation results on public benchmarks such as VisDrone-DET and TinyPerson demonstrate the robustness of the proposed method. Most impressively, the proposed HRFNet achieves a mAP of 51.0 on VisDrone-DET with 29.3 M parameters, which outperforms the extant state-of-the-art detectors. HRFNet also performs excellently in complex scenarios captured by drones, achieving the best performance on the CS-Drone dataset we built.展开更多
To improve small object detection and trajectory estimation from an aerial moving perspective,we propose the Aerial View Attention-PRB(AVA-PRB)model.AVA-PRB integrates two attention mechanisms—Coordinate Attention(CA...To improve small object detection and trajectory estimation from an aerial moving perspective,we propose the Aerial View Attention-PRB(AVA-PRB)model.AVA-PRB integrates two attention mechanisms—Coordinate Attention(CA)and the Convolutional Block Attention Module(CBAM)—to enhance detection accuracy.Additionally,Shape-IoU is employed as the loss function to refine localization precision.Our model further incorporates an adaptive feature fusion mechanism,which optimizes multi-scale object representation,ensuring robust tracking in complex aerial environments.We evaluate the performance of AVA-PRB on two benchmark datasets:Aerial Person Detection and VisDrone2019-Det.The model achieves 60.9%mAP@0.5 on the Aerial Person Detection dataset,and 51.2%mAP@0.5 on VisDrone2019-Det,demonstrating its effectiveness in aerial object detection.Beyond detection,we propose a novel trajectory estimation method that improves movement path prediction under aerial motion.Experimental results indicate that our approach reduces path deviation by up to 64%,effectively mitigating errors caused by rapid camera movements and background variations.By optimizing feature extraction and enhancing spatialtemporal coherence,our method significantly improves object tracking under aerial moving perspectives.This research addresses the limitations of fixed-camera tracking,enhancing flexibility and accuracy in aerial tracking applications.The proposed approach has broad potential for real-world applications,including surveillance,traffic monitoring,and environmental observation.展开更多
Transorbital craniocerebral injury is a relatively rare type of penetrating head injury that poses a significant threat to the ocular and cerebral structures.^([1])The clinical prognosis of transorbital craniocerebral...Transorbital craniocerebral injury is a relatively rare type of penetrating head injury that poses a significant threat to the ocular and cerebral structures.^([1])The clinical prognosis of transorbital craniocerebral injury is closely related to the size,shape,speed,nature,and trajectory of the foreign object,as well as the incidence of central nervous system damage and secondary complications.The foreign objects reported to have caused these injuries are categorized into wooden items,metallic items,^([2-8])and other materials,which penetrate the intracranial region via fi ve major pathways,including the orbital roof (OR),superior orbital fissure (SOF),inferior orbital fissure(IOF),optic canal (OC),and sphenoid wing.Herein,we present eight cases of transorbital craniocerebral injury caused by an unusual metallic foreign body.展开更多
In this paper,a two-stage light detection and ranging(LiDAR) three-dimensional(3D) object detection framework is presented,namely point-voxel dual transformer(PV-DT3D),which is a transformer-based method.In the propos...In this paper,a two-stage light detection and ranging(LiDAR) three-dimensional(3D) object detection framework is presented,namely point-voxel dual transformer(PV-DT3D),which is a transformer-based method.In the proposed PV-DT3D,point-voxel fusion features are used for proposal refinement.Specifically,keypoints are sampled from entire point cloud scene and used to encode representative scene features via a proposal-aware voxel set abstraction module.Subsequently,following the generation of proposals by the region proposal networks(RPN),the internal encoded keypoints are fed into the dual transformer encoder-decoder architecture.In 3D object detection,the proposed PV-DT3D takes advantage of both point-wise transformer and channel-wise architecture to capture contextual information from the spatial and channel dimensions.Experiments conducted on the highly competitive KITTI 3D car detection leaderboard show that the PV-DT3D achieves superior detection accuracy among state-of-the-art point-voxel-based methods.展开更多
Top-view fisheye cameras are widely used in personnel surveillance for their broad field of view,but their unique imaging characteristics pose challenges like distortion,complex scenes,scale variations,and small objec...Top-view fisheye cameras are widely used in personnel surveillance for their broad field of view,but their unique imaging characteristics pose challenges like distortion,complex scenes,scale variations,and small objects near image edges.To tackle these,we proposed peripheral focus you only look once(PF-YOLO),an enhanced YOLOv8n-based method.Firstly,we introduced a cutting-patch data augmentation strategy to mitigate the problem of insufficient small-object samples in various scenes.Secondly,to enhance the model's focus on small objects near the edges,we designed the peripheral focus loss,which uses dynamic focus coefficients to provide greater gradient gains for these objects,improving their regression accuracy.Finally,we designed the three dimensional(3D)spatial-channel coordinate attention C2f module,enhancing spatial and channel perception,suppressing noise,and improving personnel detection.Experimental results demonstrate that PF-YOLO achieves strong performance on the challenging events for person detection from overhead fisheye images(CEPDTOF)and in-the-wild events for people detection and tracking from overhead fisheye cameras(WEPDTOF)datasets.Compared to the original YOLOv8n model,PFYOLO achieves improvements on CEPDTOF with increases of 2.1%,1.7%and 2.9%in mean average precision 50(mAP 50),mAP 50-95,and tively.On WEPDTOF,PF-YOLO achieves substantial improvements with increases of 31.4%,14.9%,61.1%and 21.0%in 91.2%and 57.2%,respectively.展开更多
Aiming at the problems of low detection accuracy and large model size of existing object detection algorithms applied to complex road scenes,an improved you only look once version 8(YOLOv8)object detection algorithm f...Aiming at the problems of low detection accuracy and large model size of existing object detection algorithms applied to complex road scenes,an improved you only look once version 8(YOLOv8)object detection algorithm for infrared images,F-YOLOv8,is proposed.First,a spatial-to-depth network replaces the traditional backbone network's strided convolution or pooling layer.At the same time,it combines with the channel attention mechanism so that the neural network focuses on the channels with large weight values to better extract low-resolution image feature information;then an improved feature pyramid network of lightweight bidirectional feature pyramid network(L-BiFPN)is proposed,which can efficiently fuse features of different scales.In addition,a loss function of insertion of union based on the minimum point distance(MPDIoU)is introduced for bounding box regression,which obtains faster convergence speed and more accurate regression results.Experimental results on the FLIR dataset show that the improved algorithm can accurately detect infrared road targets in real time with 3%and 2.2%enhancement in mean average precision at 50%IoU(mAP50)and mean average precision at 50%—95%IoU(mAP50-95),respectively,and 38.1%,37.3%and 16.9%reduction in the number of model parameters,the model weight,and floating-point operations per second(FLOPs),respectively.To further demonstrate the detection capability of the improved algorithm,it is tested on the public dataset PASCAL VOC,and the results show that F-YOLO has excellent generalized detection performance.展开更多
基金funded by Princess Nourah bint Abdulrahman University Researchers Supporting Project number(PNURSP2025R410),Princess Nourah bint Abdulrahman University,Riyadh,Saudi Arabia.
文摘Human object detection and recognition is essential for elderly monitoring and assisted living however,models relying solely on pose or scene context often struggle in cluttered or visually ambiguous settings.To address this,we present SCENET-3D,a transformer-drivenmultimodal framework that unifies human-centric skeleton features with scene-object semantics for intelligent robotic vision through a three-stage pipeline.In the first stage,scene analysis,rich geometric and texture descriptors are extracted from RGB frames,including surface-normal histograms,angles between neighboring normals,Zernike moments,directional standard deviation,and Gabor-filter responses.In the second stage,scene-object analysis,non-human objects are segmented and represented using local feature descriptors and complementary surface-normal information.In the third stage,human-pose estimation,silhouettes are processed through an enhanced MoveNet to obtain 2D anatomical keypoints,which are fused with depth information and converted into RGB-based point clouds to construct pseudo-3D skeletons.Features from all three stages are fused and fed in a transformer encoder with multi-head attention to resolve visually similar activities.Experiments on UCLA(95.8%),ETRI-Activity3D(89.4%),andCAD-120(91.2%)demonstrate that combining pseudo-3D skeletonswith rich scene-object fusion significantly improves generalizable activity recognition,enabling safer elderly care,natural human–robot interaction,and robust context-aware robotic perception in real-world environments.
基金supported by the National Natural Science Foundation of China(No.62276204)the Fundamental Research Funds for the Central Universities,China(No.YJSJ24011)+1 种基金the Natural Science Basic Research Program of Shaanxi,China(Nos.2022JM-340 and 2023-JC-QN-0710)the China Postdoctoral Science Foundation(Nos.2020T130494 and 2018M633470)。
文摘Visible and infrared(RGB-IR)fusion object detection plays an important role in security,disaster relief,etc.In recent years,deep-learning-based RGB-IR fusion detection methods have been developing rapidly,but still struggle to deal with the complex and changing scenarios captured by drones,mainly due to two reasons:(A)RGB-IR fusion detectors are susceptible to inferior inputs that degrade performance and stability.(B)RGB-IR fusion detectors are susceptible to redundant features that reduce accuracy and efficiency.In this paper,an innovative RGB-IR fusion detection framework based on global-local feature optimization,named GLFDet,is proposed to improve the detection performance and efficiency of drone-captured objects.The key components of GLFDet include a Global Feature Optimization(GFO)module,a Local Feature Optimization(LFO)module and a Channel Separation Fusion(CSF)module.Specifically,GFO calculates the information content of the input image from the frequency domain and optimizes the features holistically.Then,LFO dynamically selects high-value features and filters out low-value features before fusion,which significantly improves the efficiency of fusion.Finally,CSF fuses the RGB and IR features across the corresponding channels,which avoids the rearrangement of the channel relationships and enhances the model stability.Extensive experimental results show that the proposed method achieves the best performance on three popular RGB-IR datasets Drone Vehicle,VEDAI,and LLVIP.In addition,GLFDet is more lightweight than other comparable models,making it more appealing to edge devices such as drones.The code is available at https://github.com/lao chen330/GLFDet.
基金supported in part by the National Science Foundation of China(52371372)the Project of Science and Technology Commission of Shanghai Municipality,China(22JC1401400,21190780300)the 111 Project,China(D18003)
文摘Dear Editor,This letter focuses on the fact that small objects with few pixels disappear in feature maps with large receptive fields, as the network deepens, in object detection tasks. Therefore, the detection of dense small objects is challenging.
基金funded by the Hainan Province Science and Technology Special Fund under Grant ZDYF2024GXJS292.
文摘Deep learning has made significant progress in the field of oriented object detection for remote sensing images.However,existing methods still face challenges when dealing with difficult tasks such as multi-scale targets,complex backgrounds,and small objects in remote sensing.Maintaining model lightweight to address resource constraints in remote sensing scenarios while improving task completion for remote sensing tasks remains a research hotspot.Therefore,we propose an enhanced multi-scale feature extraction lightweight network EM-YOLO based on the YOLOv8s architecture,specifically optimized for the characteristics of large target scale variations,diverse orientations,and numerous small objects in remote sensing images.Our innovations lie in two main aspects:First,a dynamic snake convolution(DSC)is introduced into the backbone network to enhance the model’s feature extraction capability for oriented targets.Second,an innovative focusing-diffusion module is designed in the feature fusion neck to effectively integrate multi-scale feature information.Finally,we introduce Layer-Adaptive Sparsity for magnitude-based Pruning(LASP)method to perform lightweight network pruning to better complete tasks in resource-constrained scenarios.Experimental results on the lightweight platform Orin demonstrate that the proposed method significantly outperforms the original YOLOv8s model in oriented remote sensing object detection tasks,and achieves comparable or superior performance to state-of-the-art methods on three authoritative remote sensing datasets(DOTA v1.0,DOTA v1.5,and HRSC2016).
文摘In modern industrial production,foreign object detection in complex environments is crucial to ensure product quality and production safety.Detection systems based on deep-learning image processing algorithms often face challenges with handling high-resolution images and achieving accurate detection against complex backgrounds.To address these issues,this study employs the PatchCore unsupervised anomaly detection algorithm combined with data augmentation techniques to enhance the system’s generalization capability across varying lighting conditions,viewing angles,and object scales.The proposed method is evaluated in a complex industrial detection scenario involving the bogie of an electric multiple unit(EMU).A dataset consisting of complex backgrounds,diverse lighting conditions,and multiple viewing angles is constructed to validate the performance of the detection system in real industrial environments.Experimental results show that the proposed model achieves an average area under the receiver operating characteristic curve(AUROC)of 0.92 and an average F1 score of 0.85.Combined with data augmentation,the proposed model exhibits improvements in AUROC by 0.06 and F1 score by 0.03,demonstrating enhanced accuracy and robustness for foreign object detection in complex industrial settings.In addition,the effects of key factors on detection performance are systematically analyzed,providing practical guidance for parameter selection in real industrial applications.
基金supported in part by the by Chongqing Research Program of Basic Research and Frontier Technology under Grant CSTB2025NSCQ-GPX1309.
文摘Small object detection has been a focus of attention since the emergence of deep learning-based object detection.Although classical object detection frameworks have made significant contributions to the development of object detection,there are still many issues to be resolved in detecting small objects due to the inherent complexity and diversity of real-world visual scenes.In particular,the YOLO(You Only Look Once)series of detection models,renowned for their real-time performance,have undergone numerous adaptations aimed at improving the detection of small targets.In this survey,we summarize the state-of-the-art YOLO-based small object detection methods.This review presents a systematic categorization of YOLO-based approaches for small-object detection,organized into four methodological avenues,namely attention-based feature enhancement,detection-head optimization,loss function,and multi-scale feature fusion strategies.We then examine the principal challenges addressed by each category.Finally,we analyze the performance of thesemethods on public benchmarks and,by comparing current approaches,identify limitations and outline directions for future research.
文摘In recent years,with the rapid advancement of artificial intelligence,object detection algorithms have made significant strides in accuracy and computational efficiency.Notably,research and applications of Anchor-Free models have opened new avenues for real-time target detection in optical remote sensing images(ORSIs).However,in the realmof adversarial attacks,developing adversarial techniques tailored to Anchor-Freemodels remains challenging.Adversarial examples generated based on Anchor-Based models often exhibit poor transferability to these new model architectures.Furthermore,the growing diversity of Anchor-Free models poses additional hurdles to achieving robust transferability of adversarial attacks.This study presents an improved cross-conv-block feature fusion You Only Look Once(YOLO)architecture,meticulously engineered to facilitate the extraction ofmore comprehensive semantic features during the backpropagation process.To address the asymmetry between densely distributed objects in ORSIs and the corresponding detector outputs,a novel dense bounding box attack strategy is proposed.This approach leverages dense target bounding boxes loss in the calculation of adversarial loss functions.Furthermore,by integrating translation-invariant(TI)and momentum-iteration(MI)adversarial methodologies,the proposed framework significantly improves the transferability of adversarial attacks.Experimental results demonstrate that our method achieves superior adversarial attack performance,with adversarial transferability rates(ATR)of 67.53%on the NWPU VHR-10 dataset and 90.71%on the HRSC2016 dataset.Compared to ensemble adversarial attack and cascaded adversarial attack approaches,our method generates adversarial examples in an average of 0.64 s,representing an approximately 14.5%improvement in efficiency under equivalent conditions.
基金supported and funded by Princess Nourah bint Abdulrahman University Researchers Supporting Project number(PNURSP2025R410),Princess Nourah bint Abdulrahman University,Riyadh,Saudi Arabia.
文摘Recognising human-object interactions(HOI)is a challenging task for traditional machine learning models,including convolutional neural networks(CNNs).Existing models show limited transferability across complex datasets such as D3D-HOI and SYSU 3D HOI.The conventional architecture of CNNs restricts their ability to handle HOI scenarios with high complexity.HOI recognition requires improved feature extraction methods to overcome the current limitations in accuracy and scalability.This work proposes a Novel quantum gate-enabled hybrid CNN(QEH-CNN)for effectiveHOI recognition.Themodel enhancesCNNperformance by integrating quantumcomputing components.The framework begins with bilateral image filtering,followed bymulti-object tracking(MOT)and Felzenszwalb superpixel segmentation.A watershed algorithm refines object boundaries by cleaning merged superpixels.Feature extraction combines a histogram of oriented gradients(HOG),Global Image Statistics for Texture(GIST)descriptors,and a novel 23-joint keypoint extractionmethod using relative joint angles and joint proximitymeasures.A fuzzy optimization process refines the extracted features before feeding them into the QEH-CNNmodel.The proposed model achieves 95.06%accuracy on the 3D-D3D-HOI dataset and 97.29%on the SYSU3DHOI dataset.Theintegration of quantum computing enhances feature optimization,leading to improved accuracy and overall model efficiency.
文摘Salient object detection(SOD)models struggle to simultaneously preserve global structure,maintain sharp object boundaries,and sustain computational efficiency in complex scenes.In this study,we propose SPSALNet,a task-driven two-stage(macro–micro)architecture that restructures the SOD process around superpixel representations.In the proposed approach,a“split-and-enhance”principle,introduced to our knowledge for the first time in the SOD literature,hierarchically classifies superpixels and then applies targeted refinement only to ambiguous or error-prone regions.At the macro stage,the image is partitioned into content-adaptive superpixel regions,and each superpixel is represented by a high-dimensional region-level feature vector.These representations define a regional decomposition problem in which superpixels are assigned to three classes:background,object interior,and transition regions.Superpixel tokens interact with a global feature vector from a deep network backbone through a cross-attention module and are projected into an enriched embedding space that jointly encodes local topology and global context.At the micro stage,the model employs a U-Net-based refinement process that allocates computational resources only to ambiguous transition regions.The image and distance–similarity maps derived from superpixels are processed through a dual-encoder pathway.Subsequently,channel-aware fusion blocks adaptively combine information from these two sources,producing sharper and more stable object boundaries.Experimental results show that SPSALNet achieves high accuracy with lower computational cost compared to recent competing methods.On the PASCAL-S and DUT-OMRON datasets,SPSALNet exhibits a clear performance advantage across all key metrics,and it ranks first on accuracy-oriented measures on HKU-IS.On the challenging DUT-OMRON benchmark,SPSALNet reaches a MAE of 0.034.Across all datasets,it preserves object boundaries and regional structure in a stable and competitive manner.
基金funded by the National Natural Science Foundation of China under Grant No.62371187the Open Program of Hunan Intelligent Rehabilitation Robot and Auxiliary Equipment Engineering Technology Research Center under Grant No.2024JS101.
文摘The ubiquity of mobile devices has driven advancements in mobile object detection.However,challenges in multi-scale object detection in open,complex environments persist due to limited computational resources.Traditional approaches like network compression,quantization,and lightweight design often sacrifice accuracy or feature representation robustness.This article introduces the Fast Multi-scale Channel Shuffling Network(FMCSNet),a novel lightweight detection model optimized for mobile devices.FMCSNet integrates a fully convolutional Multilayer Perceptron(MLP)module,offering global perception without significantly increasing parameters,effectively bridging the gap between CNNs and Vision Transformers.FMCSNet achieves a delicate balance between computation and accuracy mainly by two key modules:the ShiftMLP module,including a shift operation and an MLP module,and a Partial group Convolutional(PGConv)module,reducing computation while enhancing information exchange between channels.With a computational complexity of 1.4G FLOPs and 1.3M parameters,FMCSNet outperforms CNN-based and DWConv-based ShuffleNetv2 by 1%and 4.5%mAP on the Pascal VOC 2007 dataset,respectively.Additionally,FMCSNet achieves a mAP of 30.0(0.5:0.95 IoU threshold)with only 2.5G FLOPs and 2.0M parameters.It achieves 32 FPS on low-performance i5-series CPUs,meeting real-time detection requirements.The versatility of the PGConv module’s adaptability across scenarios further highlights FMCSNet as a promising solution for real-time mobile object detection.
基金supported by the confidential research grant No.a8317。
文摘To address the issues of frequent identity switches(IDs)and degraded identification accuracy in multi object tracking(MOT)under complex occlusion scenarios,this study proposes an occlusion-robust tracking framework based on face-pedestrian joint feature modeling.By constructing a joint tracking model centered on“intra-class independent tracking+cross-category dynamic binding”,designing a multi-modal matching metric with spatio-temporal and appearance constraints,and innovatively introducing a cross-category feature mutual verification mechanism and a dual matching strategy,this work effectively resolves performance degradation in traditional single-category tracking methods caused by short-term occlusion,cross-camera tracking,and crowded environments.Experiments on the Chokepoint_Face_Pedestrian_Track test set demonstrate that in complex scenes,the proposed method improves Face-Pedestrian Matching F1 area under the curve(F1 AUC)by approximately 4 to 43 percentage points compared to several traditional methods.The joint tracking model achieves overall performance metrics of IDF1:85.1825%and MOTA:86.5956%,representing improvements of 0.91 and 0.06 percentage points,respectively,over the baseline model.Ablation studies confirm the effectiveness of key modules such as the Intersection over Area(IoA)/Intersection over Union(IoU)joint metric and dynamic threshold adjustment,validating the significant role of the cross-category identity matching mechanism in enhancing tracking stability.Our_model shows a 16.7%frame per second(FPS)drop vs.fairness of detection and re-identification in multiple object tracking(FairMOT),with its cross-category binding module adding aboute 10%overhead,yet maintains near-real-time performance for essential face-pedestrian tracking at small resolutions.
文摘With the rapid expansion of drone applications,accurate detection of objects in aerial imagery has become crucial for intelligent transportation,urban management,and emergency rescue missions.However,existing methods face numerous challenges in practical deployment,including scale variation handling,feature degradation,and complex backgrounds.To address these issues,we propose Edge-enhanced and Detail-Capturing You Only Look Once(EHDC-YOLO),a novel framework for object detection in Unmanned Aerial Vehicle(UAV)imagery.Based on the You Only Look Once version 11 nano(YOLOv11n)baseline,EHDC-YOLO systematically introduces several architectural enhancements:(1)a Multi-Scale Edge Enhancement(MSEE)module that leverages multi-scale pooling and edge information to enhance boundary feature extraction;(2)an Enhanced Feature Pyramid Network(EFPN)that integrates P2-level features with Cross Stage Partial(CSP)structures and OmniKernel convolutions for better fine-grained representation;and(3)Dynamic Head(DyHead)with multi-dimensional attention mechanisms for enhanced cross-scale modeling and perspective adaptability.Comprehensive experiments on the Vision meets Drones for Detection(VisDrone-DET)2019 dataset demonstrate that EHDC-YOLO achieves significant improvements,increasing mean Average Precision(mAP)@0.5 from 33.2%to 46.1%(an absolute improvement of 12.9 percentage points)and mAP@0.5:0.95 from 19.5%to 28.0%(an absolute improvement of 8.5 percentage points)compared with the YOLOv11n baseline,while maintaining a reasonable parameter count(2.81 M vs the baseline’s 2.58 M).Further ablation studies confirm the effectiveness of each proposed component,while visualization results highlight EHDC-YOLO’s superior performance in detecting objects and handling occlusions in complex drone scenarios.
文摘Modern manufacturing processes have become more reliant on automation because of the accelerated transition from Industry 3.0 to Industry 4.0.Manual inspection of products on assembly lines remains inefficient,prone to errors and lacks consistency,emphasizing the need for a reliable and automated inspection system.Leveraging both object detection and image segmentation approaches,this research proposes a vision-based solution for the detection of various kinds of tools in the toolkit using deep learning(DL)models.Two Intel RealSense D455f depth cameras were arranged in a top down configuration to capture both RGB and depth images of the toolkits.After applying multiple constraints and enhancing them through preprocessing and augmentation,a dataset consisting of 3300 annotated RGB-D photos was generated.Several DL models were selected through a comprehensive assessment of mean Average Precision(mAP),precision-recall equilibrium,inference latency(target≥30 FPS),and computational burden,resulting in a preference for YOLO and Region-based Convolutional Neural Networks(R-CNN)variants over ViT-based models due to the latter’s increased latency and resource requirements.YOLOV5,YOLOV8,YOLOV11,Faster R-CNN,and Mask R-CNN were trained on the annotated dataset and evaluated using key performance metrics(Recall,Accuracy,F1-score,and Precision).YOLOV11 demonstrated balanced excellence with 93.0%precision,89.9%recall,and a 90.6%F1-score in object detection,as well as 96.9%precision,95.3%recall,and a 96.5%F1-score in instance segmentation with an average inference time of 25 ms per frame(≈40 FPS),demonstrating real-time performance.Leveraging these results,a YOLOV11-based windows application was successfully deployed in a real-time assembly line environment,where it accurately processed live video streams to detect and segment tools within toolkits,demonstrating its practical effectiveness in industrial automation.The application is capable of precisely measuring socket dimensions by utilising edge detection techniques on YOLOv11 segmentation masks,in addition to detection and segmentation.This makes it possible to do specification-level quality control right on the assembly line,which improves the ability to examine things in real time.The implementation is a big step forward for intelligent manufacturing in the Industry 4.0 paradigm.It provides a scalable,efficient,and accurate way to do automated inspection and dimensional verification activities.
文摘The goal of the present work is to demonstrate the potential of Artificial Neural Network(ANN)-driven Genetic Algorithm(GA)methods for energy efficiency and economic performance optimization of energy efficiency measures in a multi-family house building in Greece.The energy efficiency measures include different heating/cooling systems(such as low-temperature and high-temperature heat pumps,natural gas boilers,split units),building envelope components for floor,walls,roof and windows of variable heat transfer coefficients,the installation of solar thermal collectors and PVs.The calculations of the building loads and investment and operating and maintenance costs of the measures are based on the methodology defined in Directive 2010/31/EU,while economic assumptions are based on EN 15459-1 standard.Typically,multi-objective optimization of energy efficiency measures often requires the simulation of very large numbers of cases involving numerous possible combinations,resulting in intense computational load.The results of the study indicate that ANN-driven GA methods can be used as an alternative,valuable tool for reliably predicting the optimal measures which minimize primary energy consumption and life cycle cost of the building with greatly reduced computational requirements.Through GA methods,the computational time needed for obtaining the optimal solutions is reduced by 96.4%-96.8%.
基金supported by the National Natural Science Foundation of China(Nos.62276204 and 62203343)the Fundamental Research Funds for the Central Universities(No.YJSJ24011)+1 种基金the Natural Science Basic Research Program of Shanxi,China(Nos.2022JM-340 and 2023-JC-QN-0710)the China Postdoctoral Science Foundation(Nos.2020T130494 and 2018M633470).
文摘Drone-based small object detection is of great significance in practical applications such as military actions, disaster rescue, transportation, etc. However, the severe scale differences in objects captured by drones and lack of detail information for small-scale objects make drone-based small object detection a formidable challenge. To address these issues, we first develop a mathematical model to explore how changing receptive fields impacts the polynomial fitting results. Subsequently, based on the obtained conclusions, we propose a simple but effective Hybrid Receptive Field Network (HRFNet), whose modules include Hybrid Feature Augmentation (HFA), Hybrid Feature Pyramid (HFP) and Dual Scale Head (DSH). Specifically, HFA employs parallel dilated convolution kernels of different sizes to extend shallow features with different receptive fields, committed to improving the multi-scale adaptability of the network;HFP enhances the perception of small objects by capturing contextual information across layers, while DSH reconstructs the original prediction head utilizing a set of high-resolution features and ultrahigh-resolution features. In addition, in order to train HRFNet, the corresponding dual-scale loss function is designed. Finally, comprehensive evaluation results on public benchmarks such as VisDrone-DET and TinyPerson demonstrate the robustness of the proposed method. Most impressively, the proposed HRFNet achieves a mAP of 51.0 on VisDrone-DET with 29.3 M parameters, which outperforms the extant state-of-the-art detectors. HRFNet also performs excellently in complex scenarios captured by drones, achieving the best performance on the CS-Drone dataset we built.
基金funded by theNational Science and TechnologyCouncil(NSTC),Taiwan,under grant numbers NSTC 113-2634-F-A49-007 and NSTC 112-2634-F-A49-007.
文摘To improve small object detection and trajectory estimation from an aerial moving perspective,we propose the Aerial View Attention-PRB(AVA-PRB)model.AVA-PRB integrates two attention mechanisms—Coordinate Attention(CA)and the Convolutional Block Attention Module(CBAM)—to enhance detection accuracy.Additionally,Shape-IoU is employed as the loss function to refine localization precision.Our model further incorporates an adaptive feature fusion mechanism,which optimizes multi-scale object representation,ensuring robust tracking in complex aerial environments.We evaluate the performance of AVA-PRB on two benchmark datasets:Aerial Person Detection and VisDrone2019-Det.The model achieves 60.9%mAP@0.5 on the Aerial Person Detection dataset,and 51.2%mAP@0.5 on VisDrone2019-Det,demonstrating its effectiveness in aerial object detection.Beyond detection,we propose a novel trajectory estimation method that improves movement path prediction under aerial motion.Experimental results indicate that our approach reduces path deviation by up to 64%,effectively mitigating errors caused by rapid camera movements and background variations.By optimizing feature extraction and enhancing spatialtemporal coherence,our method significantly improves object tracking under aerial moving perspectives.This research addresses the limitations of fixed-camera tracking,enhancing flexibility and accuracy in aerial tracking applications.The proposed approach has broad potential for real-world applications,including surveillance,traffic monitoring,and environmental observation.
文摘Transorbital craniocerebral injury is a relatively rare type of penetrating head injury that poses a significant threat to the ocular and cerebral structures.^([1])The clinical prognosis of transorbital craniocerebral injury is closely related to the size,shape,speed,nature,and trajectory of the foreign object,as well as the incidence of central nervous system damage and secondary complications.The foreign objects reported to have caused these injuries are categorized into wooden items,metallic items,^([2-8])and other materials,which penetrate the intracranial region via fi ve major pathways,including the orbital roof (OR),superior orbital fissure (SOF),inferior orbital fissure(IOF),optic canal (OC),and sphenoid wing.Herein,we present eight cases of transorbital craniocerebral injury caused by an unusual metallic foreign body.
基金supported by the Natural Science Foundation of China (No.62103298)the South African National Research Foundation (Nos.132797 and 137951)。
文摘In this paper,a two-stage light detection and ranging(LiDAR) three-dimensional(3D) object detection framework is presented,namely point-voxel dual transformer(PV-DT3D),which is a transformer-based method.In the proposed PV-DT3D,point-voxel fusion features are used for proposal refinement.Specifically,keypoints are sampled from entire point cloud scene and used to encode representative scene features via a proposal-aware voxel set abstraction module.Subsequently,following the generation of proposals by the region proposal networks(RPN),the internal encoded keypoints are fed into the dual transformer encoder-decoder architecture.In 3D object detection,the proposed PV-DT3D takes advantage of both point-wise transformer and channel-wise architecture to capture contextual information from the spatial and channel dimensions.Experiments conducted on the highly competitive KITTI 3D car detection leaderboard show that the PV-DT3D achieves superior detection accuracy among state-of-the-art point-voxel-based methods.
基金supported by National Natural Science Foundation of China(Nos.62171042,62102033,U24A20331)the R&D Program of Beijing Municipal Education Commission(No.KZ202211417048)+2 种基金the Project of Construction and Support for High-Level Innovative Teams of Beijing Municipal Institutions(No.BPHR20220121)Beijing Natural Science Foundation(Nos.4232026,4242020)the Academic Research Projects of Beijing Union University(Nos.ZKZD202302,ZK20202403)。
文摘Top-view fisheye cameras are widely used in personnel surveillance for their broad field of view,but their unique imaging characteristics pose challenges like distortion,complex scenes,scale variations,and small objects near image edges.To tackle these,we proposed peripheral focus you only look once(PF-YOLO),an enhanced YOLOv8n-based method.Firstly,we introduced a cutting-patch data augmentation strategy to mitigate the problem of insufficient small-object samples in various scenes.Secondly,to enhance the model's focus on small objects near the edges,we designed the peripheral focus loss,which uses dynamic focus coefficients to provide greater gradient gains for these objects,improving their regression accuracy.Finally,we designed the three dimensional(3D)spatial-channel coordinate attention C2f module,enhancing spatial and channel perception,suppressing noise,and improving personnel detection.Experimental results demonstrate that PF-YOLO achieves strong performance on the challenging events for person detection from overhead fisheye images(CEPDTOF)and in-the-wild events for people detection and tracking from overhead fisheye cameras(WEPDTOF)datasets.Compared to the original YOLOv8n model,PFYOLO achieves improvements on CEPDTOF with increases of 2.1%,1.7%and 2.9%in mean average precision 50(mAP 50),mAP 50-95,and tively.On WEPDTOF,PF-YOLO achieves substantial improvements with increases of 31.4%,14.9%,61.1%and 21.0%in 91.2%and 57.2%,respectively.
基金supported by the National Natural Science Foundation of China(No.62103298)。
文摘Aiming at the problems of low detection accuracy and large model size of existing object detection algorithms applied to complex road scenes,an improved you only look once version 8(YOLOv8)object detection algorithm for infrared images,F-YOLOv8,is proposed.First,a spatial-to-depth network replaces the traditional backbone network's strided convolution or pooling layer.At the same time,it combines with the channel attention mechanism so that the neural network focuses on the channels with large weight values to better extract low-resolution image feature information;then an improved feature pyramid network of lightweight bidirectional feature pyramid network(L-BiFPN)is proposed,which can efficiently fuse features of different scales.In addition,a loss function of insertion of union based on the minimum point distance(MPDIoU)is introduced for bounding box regression,which obtains faster convergence speed and more accurate regression results.Experimental results on the FLIR dataset show that the improved algorithm can accurately detect infrared road targets in real time with 3%and 2.2%enhancement in mean average precision at 50%IoU(mAP50)and mean average precision at 50%—95%IoU(mAP50-95),respectively,and 38.1%,37.3%and 16.9%reduction in the number of model parameters,the model weight,and floating-point operations per second(FLOPs),respectively.To further demonstrate the detection capability of the improved algorithm,it is tested on the public dataset PASCAL VOC,and the results show that F-YOLO has excellent generalized detection performance.
