计算机科学 ›› 2025, Vol. 52 ›› Issue (11): 184-195.doi: 10.11896/jsjkx.241100107
王晓峰, 黄俊俊, 谭文雅, 沈紫璇
WANG Xiaofeng, HUANG Junjun, TAN Wenya, SHEN Zixuan
摘要： 在深度网络的前馈过程中,输入数据的特征信息会被抽象和压缩,导致部分对于目标检测关键的特征信息被弱化。基于YOLOv11n,提出了深度特征强化与路径聚合优化的目标检测方法。首先,设计全局-局部特征增强模块GLFEM(Global-Local Feature Enhancement Module),结合特征图局部特征与全局特征,强化深层网络特征的表达能力。然后,设计自适应特征增强模块AFEM(Adaptive Feature Enhancement Module),根据特征的可靠性动态增强深层网络的特征提取能力。最后,对路径聚合特征金字塔网络进行优化,融合了不同层次之间的特征信息,减少了层次之间的语义信息差。在VisDrone,NWPU VHR-10和TinyPerson这3个公共数据集上的实验结果表明,该方法的平均检测精度相较于当前先进的目标检测器均有所提升。在自建数据集AirportTiny上进行实验,该方法同样取得了不错的效果,具有良好的泛化能力。
中图分类号:
| [1]CHEN C,QI J,LIU X,et al.Weakly Misalignment-free Adap-tive Feature Alignment for UAVs-based Multimodal Object Detection[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2024:26836-26845. [2]SOBEK J,MEDINA INOJOSA J R,MEDINA INOJOSA B J,et al.MedYOLO:A Medical Image Object Detection Framework [J].Journal of Imaging Informatics in Medicine,2024,37:3208-3216. [3]WANG Q,LIU F,ZOU R,et al.Enhancing medical image object detection with collaborative multi-agent deep Q-networks and multi-scale representation [J].EURASIP Journal on Advances in Signal Processing,2023,2023(1):132. [4]XU Q,LIN X,CAI M,et al.End-to-End Joint Multi-Object De-tection and Tracking for Intelligent Transportation Systems [J].Chinese Journal of Mechanical Engineering,2023,36(1):138. [5]ZHAO R,TANG S,SUPENI E E B,et al.A Review of Object Detection in Traffic Scenes Based on Deep Learning [J].Applied Mathematics and Nonlinear Sciences,2023,9(1):1-25. [6]SARACENI L,MOTOI I M,NARDI D,et al.AgriSORT:ASimple Online Real-time Tracking-by-Detection framework for robotics in precision agriculture[C]//Proceedings of the 2024 IEEE International Conference on Robotics and Automation(ICRA).IEEE,2024:2675-2682. [7]ZHAO P,ZHOU W,NA L.High-precision object detection network for automate pear picking [J].Scientific Reports,2024,14(1):14965. [8]MUJKIC E,CHRISTIANSEN M P,RAVN O.Object Detection for Agricultural Vehicles:Ensemble Method Based on Hierarchy of Classes [J].Sensors,2023,23(16):7285. [9]ZHAO Y,LYU W,XU S,et al.Detrs beat yolos on real-time object detection[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2024:16965-16974. [10]XIAO J S,ZHAO T,ZHOU J,et al.Small Target DetectionNetwork Based on Context Augmentation and Feature Refinement[J].Journal of Computer Research and Development,2023,60(2):465-474. [11]JIANG Z T,ZHAI F S,QIAN Y,et al.Low Illumination Object Detection Combined with Feature Enhancement and Multi-Scale Receptive Field[J].Journal of Computer Research and Development,2023,60(4):903-915. [12]ZHANG K H,SHEN H K.Solder joint defect detection in the connectors using improved Faster- RCNN algorithm[J].Applied Sciences,2021,11(2):576. [13]YANG A M,JIANG T Y,HAN Y,et al.Research on application of on-line melting in-SITU visual inspection of iron ore powder based on Faster R-CNN[J].Alexandria Engineering Journal,2022,61(11):8963-8971. [14]KUMAR A,MANIKANDAN R.Brain tumor detection usingdeep neural network- based classifier[C]//Proceedings of the 2022 International Conference on Innovative Computing and Communications.Singapore:Springer,2022:173-181. [15]TERVEN J,CÓRDOVA-ESPARZA D M,ROMERO-GONZÁ-LEZ J A.A comprehensive review of yolo architectures in computer vision:From yolov1 to yolov8 and yolo-nas [J].Machine Learning and Knowledge Extraction,2023,5(4):1680-1716. [16]SAPKOTA R,MENG Z C,CHURUVIJA M,et al.Comprehensive Performance Evaluation of YOLO11,YOLOv10,YOLOv9 and YOLOv8 on Detecting and Counting Fruitlet in Complex Orchard Environments[J].arXiv:2407.12040,2024. [17]WANG C Y,BOCHKOVSKIY A,LIAO H Y M.YOLOv7:Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors[C]//Proceedings of the IEEE/CVF Confe-rence on Computer Vision and Pattern Recognition.2023:7464-7475. [18]HE Z,WANG K,FANG T,et al.Comprehensive Performance Evaluation of YOLOv11,YOLOv10,YOLOv9,YOLOv8 and YOLOv5 on Object Detection of Power Equipment [J].arXiv:2411.18871,2024. [19]ZHANG Y,XIA Y.Object Detection Method with Multi-scaleFeature Fusion for R-emote Sensing Images[J].Computer Science,2024,51(3):165-173. [20]TERVEN J,CÓRDOVA-ESPARZA D M,ROMERO-GONZÁ-LEZ J A.A comprehensive review of yolo architectures in computer vision:From YOLOv1 to YOLOv8 and YOLO-nas [J].Machine Learning and Knowledge Extraction,2023,5(4):1680-1716. [21]WANG C Y,YEH I H,LIAO H Y.Yolov9:Learning what you want to learn using programmable gradient information [C]//Proceedings of the European Conference on Computer Vision.Cham:Springer.2025:1-21. [22]LIU W,ANGUELOV D,ERHAN D,et al.Ssd:Single shotmultibox detector[C]//Proceedings of the Computer Vision-ECCV 2016:14th European Conference.Springer.2016:21-37. [23]QUE Y,GAN M H,LIU Z W.Object Detection with Receptive Field Expansion and Multi-branch Aggregation[EB/OL].https://doi.org/10.11896/jsjkx.230600151. [24]LI Y C,ZHANG R,WANG J B,et al.Re-parameter-ization Enhanced Dual-modal Realtime Object Detection Model[J].Computer Science,2024,51(9):162-172. [25]WANG J,CHEN Y,DONG Z,et al.Improved YOLOv5 network for real-time multi-scale traffic sign detection [J].Neural Computing and Applications,2023,35(10):7853-7865. [26]NI J,ZHU S,TANG G,et al.A small-object detection modelbased on improved YOLOv8s for UAV image scenarios [J].Remote Sensing,2024,16(13):2465. [27]WEI J,NI L,LUO L,et al.GFS-YOLO11:A Maturity Detection Model for Multi-Variety Tomato [J].Agronomy,2024,14(11):2644. [28]JOOSHIN H K,NANGIR M,SEYEDARABI H.Inception-YOLO:Computational cost and accuracy improvement of the YOLOv5 model based on employing modified CSP,SPPF,and inception modules [J].IET Image Processing,2024,18(8):1985-1999. [29]HE K,ZHANG X,REN S,et al.Deep residual learning forimage recognition [C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2016:770-778. [30]CAGNETTA F,PETRINI L,TOMASINI U M,et al.How deep neural networks learn compositional data:The random hierarchy model [J].Physical Review X,2024,14(3):031001. [31]LIU M,WANG H,DU L,et al.Bearing-detr:A lightweightdeep learning model for bearing defect detection based on RT-DETR [J].Sensors,2024,24(13):4262. [32]LA MALFA E,LA MALFA G,NICOSIA G,et al.Characterizing learning dynamics of deep neural networks via complex networks[C]//Proceedings of the 2021 IEEE 33rd International Conference on Tools with Artificial Intelligence(ICTAI).IEEE,2021:344-351. [33]LIN T Y,DOLLÁR P,GIRSHICK R,et al.Feature pyramidnetworks for object detection [C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2017:2117-2125. [34]LIU S,QI L,QIN H,et al.Path aggregation network for instance segmentation[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2018:8759-8768. [35]TAN M,PANG R,LE Q V.Efficientdet:Scalable and efficient object detection[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2020:10781-10790. [36]QIU Y F,XIN H.Target Detection Algorithm Based on Global Feature Fusion in Parallel Dual Path Backbone[J].Journal of Frontiers of Computer Science and Technology,2024,18(12):3247-3259. [37]SHI Y,WANG L,YAO Y P,et al.Small Object Detection Based on Enhanced Feature Pyramid and Focal-AIoU Loss[J].Journal of Frontiers of Computer Science and Technology,2025,19(3):693-702. [38]HAN B,HE L,KE J,et al.Weighted parallel decoupled feature pyramid network for object detection [J].Neurocomputing,2024,593:127809. [39]PENG Z,HUANG W,GU S,et al.Conformer:Local featurescoupling global representations for visual recognition[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision.2021:367-376. [40]GEIGER B C,KUBIN G.Information bottleneck:Theory and applications in deep learning [J].Entropy,2020,22(12):1408. [41]LIU Z,WANG B,LI Y,et al.UnitModule:A lightweight joint image enhancement module for underwater object detection [J].Pattern Recognition,2024,151:110435. [42]HUANG G,LIU Z,VAN DER MAATEN L,et al.Densely con-nected convolutional networks[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2017:4700-4708. [43]NING Q,DONG W,LI X,et al.Uncertainty-driven loss for sin-gle image super-resolution [J].Advances in Neural Information Processing Systems,2021,34:16398-16409. |
|
||
首页