YOLO-PD：轻量级实时行人检测算法

doi:10.3778/j.issn.1002-8331.2507-0218

摘要/Abstract

摘要： 针对边缘端设备资源受限和行人检测算法在小目标物体、多尺度物体和重叠遮挡造成的几何形变物体上检测精度低的问题，提出了一种适合部署在边缘端设备的基于YOLOv8的行人检测算法YOLO-PD。针对小目标物体，设计局部Transformer块（partial Transformer block，PTB），增强模型的特征提取能力，PTB的混合结构在保持高效特征提取的同时还能降低计算成本；针对多尺度物体，设计金字塔共享空洞卷积（pyramid shared dilation convolution，PSDC），利用共享权重参数的多尺度空洞卷积来提高模型多尺度特征提取能力的同时减小模块体积；针对几何形变物体，设计轻量级可变形动态检测头（light deformable dynamic head，LDDH），在检测头中通过动态调整加权因子来提升检测精度、通过可变形卷积来更精准地获取形变物体的特征。实验结果表明，与基线模型YOLOv8n相比，YOLO-PD在自建的行人检测数据集COCO-Person和VOC-Person上的mAP50分别提高了2.9和1.9个百分点，参数量减少了34.3%；在公开的行人检测数据集WidePerson上，mAP50和mAP50：95分别提高了1.8和1.1个百分点。在行人检测任务上该算法检测精度高、泛化能力强、极小的参数量让它适合部署在边缘端设备。

关键词: YOLOv8, 行人检测, WidePerson数据集

Abstract: To address the limited resources of edge devices and the low detection accuracy of pedestrian detection algorithms on small objects, multi-scale objects, and geometrically deformed objects caused by overlapping occlusions, YOLO-PD is proposed which is a YOLOv8-based pedestrian detection algorithm suitable for edge deployment. For small objects, a partial Transformer block (PTB) is designed to enhance feature extraction. Hybrid structure of PTB maintains efficient feature extraction while reducing computational costs. For multi-scale objects, a pyramid shared dilation convolution (PSDC) is introduced, leveraging shared-weight multi-scale dilation convolutions to improve multi-scale feature extraction while minimizing module size. For geometrically deformed objects, a light deformable dynamic head (LDDH) is developed. It dynamically adjusts weighting factors to improve detection accuracy and employs deformable convolutions to better capture features of deformed objects. Experimental results show that compared to the baseline model YOLOv8n, YOLO-PD achieves 2.9 and 1.9 percentage points higher mAP50 on the custom pedestrian detection datasets COCO-Person and VOC-Person, respectively, while reducing parameters by 34.3%. On the public dataset WidePerson, it improves mAP50 by 1.8 percentage points and mAP50:95 by 1.1 percentage points. The algorithm excels in pedestrian detection with high accuracy, strong generalization, and minimal parameters, making it ideal for edge devices.

Key words: YOLOv8, pedestrian detection, WidePerson dataset

陈胜宝, 施隆照. YOLO-PD：轻量级实时行人检测算法[J]. 计算机工程与应用, 2026, 62(6): 122-133.

CHEN Shengbao, SHI Longzhao. YOLO-PD:Lightweight Real-Time Pedestrian Detection Algorithm[J]. Computer Engineering and Applications, 2026, 62(6): 122-133.

参考文献

[1] 齐鹏宇，王洪元，张继，等.基于改进FCOS的拥挤行人检测算法[J].智能系统学报，2021，16（4）：811-818.
QI P Y，WANG H Y，ZHANG J，et al.Crowded pedestrian detection algorithm based on improved FCOS[J].CAAI Transactions on Intelligent Systems，2021，16（4）：811-818.
[2] 罗艳，张重阳，田永鸿，等.深度学习行人检测方法综述[J].中国图象图形学报，2022，27（7）：2094-2111.
LUO Y，ZHANG C Y，TIAN Y H，et al.An overview of deep learning based pedestrian detection algorithms[J].Journal of Image and Graphics，2022，27（7）：2094-2111.
[3] 赵才荣，齐鼎，窦曙光，等.智能视频监控关键技术：行人再识别研究综述[J].中国科学（信息科学），2021，51（12）：1979-2015.
ZHAO C R，QI D，DOU S G，et al.Key technology for intelligent video surveillance：a review of person re-identification[J].Scientia Sinica（Informationis），2021，51（12）：1979-2015.
[4] 陈宁，李梦璐，袁皓，等.遮挡情形下的行人检测方法综述[J].计算机工程与应用，2020，56（16）：13-20.
CHEN N，LI M L，YUAN H，et al.Review of pedestrian detection with occlusion[J].Computer Engineering and Applications，2020，56（16）：13-20.
[5] DALAL N，TRIGGS B.Histograms of oriented gradients for human detection[C]//Proceedings of the 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.Piscataway：IEEE，2005：886-893.
[6] OJALA T，PIETIK?INEN M，HARWOOD D.A comparative study of texture measures with classification based on featured distributions[J].Pattern Recognition，1996，29（1）：51-59.
[7] 田仙仙，鲍泓，徐成.一种改进HOG特征的行人检测算法[J].计算机科学，2014，41（9）：320-324.
TIAN X X，BAO H，XU C.Improved HOG algorithm of pedestrian detection[J].Computer Science，2014，41（9）：320-324.
[8] PAPAGEORGIOU C，POGGIO T.A trainable system for object detection[J].International Journal of Computer Vision，2000，38（1）：15-33.
[9] GIRSHICK R，DONAHUE J，DARRELL T，et al.Rich feature hierarchies for accurate object detection and semantic segmentation[C]//Proceedings of the 2014 IEEE Conference on Computer Vision and Pattern Recognition.Piscataway：IEEE，2014：580-587.
[10] GIRSHICK R.Fast R?CNN[C]//Proceedings of the 2015 IEEE International Conference on Computer Vision.Piscataway：IEEE，2016：1440-1448.
[11] REN S Q，HE K M，GIRSHICK R，et al.Faster R-CNN：towards real-time object detection with region proposal networks[C]//Advances in Neural Information Processing Systems，2015：91-99.
[12] CAI Z W，VASCONCELOS N.Cascade R-CNN：delving into high quality object detection[C]//Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.Piscataway：IEEE，2018：6154-6162.
[13] CAO J L，CHOLAKKAL H，ANWER R M，et al.D2Det：towards high quality object detection and instance segmentation[C]//Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition.Piscataway：IEEE，2020：11482-11491.
[14] LIU W，ANGUELOV D，ERHAN D，et al.SSD：single shot MultiBox detector[C]//Proceedings of the 14th European Conference on Computer Vision.Cham：Springer，2016：21-37.
[15] TERVEN J，CóRDOVA-ESPARZA D M，ROMERO-GONZáLEZ J A，et al.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] 裴伟，许晏铭，朱永英，等.改进的SSD航拍目标检测方法[J].软件学报，2019，30（3）：738-758.
PEI W，XU Y M，ZHU Y Y，et al.The target detection method of aerial photography images with improved SSD[J].Journal of Software，2019，30（3）：738-758.
[17] 顾伟，李菲菲，陈虬.基于多特征融合的行人检测方法[J].电子科技，2021，34（5）：29-34.
GU W，LI F F，CHEN Q.Pedestrian detection algorithm based on multiple feature fusion[J].Electronic Science and Technology，2021，34（5）：29-34.
[18] ZHAO A X，ZHAO Y K，ZHENG Q H，et al.Robust pedestrian detection and intrusion judgment in coal yard hazard areas via 3D LiDAR-based deep learning[J].Sensors，2025，25（18）：5908.
[19] MA X M，LI N，WU D，et al.Multi-scale pedestrian detection algorithm based on joint head and overall information[J].Chinese Journal of Liquid Crystals and Displays，2025，40（6）：931-941.
[20] QU Z H，LIU H Y，KONG W G，et al.LP-YOLO：an improved lightweight pedestrian detection algorithm based on YOLOv11[J].Digital Signal Processing，2025，165：105343.
[21] REDMON J，DIVVALA S，GIRSHICK R，et al.You only look once：unified，real-time object detection[C]//Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition.Piscataway：IEEE，2016：779-788.
[22] REDMON J，FARHADI A.YOLOv3：an incremental improvement[EB/OL].[2025-05-18].https://arxiv.org/abs/1804.
02767.
[23] LI C Y，LI L L，JIANG H L，et al.YOLOv6：a single-stage object detection framework for industrial applications[EB/OL].[2025-05-18].https://arxiv.org/abs/2209.02976.
[24] WANG C Y，BOCHKOVSKIY A，LIAO H M.YOLOv7：trainable bag-of-freebies sets new state-of-the-art for real-time object detectors[EB/OL].[2025-05-18].https://arxiv.org/abs/2207.02696.
[25] WANG A，CHEN H，LIU L H，et al.YOLOv10：real-time end-to-end object detection[EB/OL].[2025-05-20].https://arxiv.org/abs/2405.14458.
[26] LIU H J，LIU F Q，FAN X Y，et al.Polarized self-attention：towards high-quality pixel-wise regression[EB/OL].[2025-05-22].https://arxiv.org/abs/2107.00782.
[27] SHI D.TransNeXt：robust foveal visual perception for vision transformers[C]//Proceedings of the 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition.Piscataway：IEEE，2024：17773-17783.
[28] WU Y X，HE K M.Group normalization[J].International Journal of Computer Vision，2020，128（3）：742-755.
[29] TIAN Z，SHEN C H，CHEN H，et al.FCOS：fully convolutional one-stage object detection[C]//Proceedings of the 2019 IEEE/CVF International Conference on Computer Vision.Piscataway：IEEE，2020：9626-9635.
[30] ZHU X Z，HU H，LIN S，et al.Deformable ConvNets V2：more deformable，better results[C]//Proceedings of the 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition.Piscataway：IEEE，2020：9300-9308.
[31] 胡倩，皮建勇，胡伟超，等.基于改进YOLOv5的密集行人检测算法[J].计算机工程，2025，51（3）：216-228.
HU Q，PI J Y，HU W C，et al.Dense pedestrian detection algorithm based on improved YOLOv5[J].Computer Engineering，2025，51（3）：216-228.
[32] 黄昆，齐肇建，王娟敏，等.基于改进YOLOv8的密集行人检测模型[J].计算机工程，2025，51（5）：133-142.
HUANG K，QI Z J，WANG J M，et al.Aggregation pedestrian detection model based on improved YOLOv8[J].Computer Engineering，2025，51（5）：133-142.
[33] 汤静雯，赖惠成，王同官.远距离情形下的改进YOLOv8行人检测算法[J].计算机工程，2025，51（4）：303-313.
TANG J W，LAI H C，WANG T G.Improved YOLOv8 pedestrian detection algorithm for long-distance situations[J].Computer Engineering，2025，51（4）：303-313.
[34] OUSSOUADDI M，BOUAZIZI O，EL MOURABIT A，et al.DSR?YOLO：a lightweight and efficient YOLOv8 model for enhanced pedestrian detection[J].Cognitive Robotics，2025，5：152-165.
[35] 王泽宇，徐慧英，朱信忠，等.基于YOLOv8改进的密集行人检测算法：MER-YOLO[J].计算机工程与科学，2024，46（6）：1050-1062.
WANG Z Y，XU H Y，ZHU X Z，et al.An improved dense pedestrian detection algorithm based on YOLOv8：MER-YOLO[J].Computer Engineering & Science，2024，46（6）：1050-1062.
[36] TSAI C Y，WANG R Y，CHIU Y C.SW-YOLOX：a YOLOX-based real-time pedestrian detector with shift window-mixed attention mechanism[J].Neurocomputing，2024，606：128357.