| 83 | 0 | 154 |
| 下载次数 | 被引频次 | 阅读次数 |
由于行人姿态变化的多样性、背景环境的复杂性等因素的干扰，会导致行人特征的表达能力不足，进而影响到行人重识别的准确性。本文基于深度学习的AlignedReID++模型进行改进，将DenseNet121网络与AlignedReID++模型的主干网络ResNet50融合，利用两者的优点，增强模型的特征提取能力；通过引入跨维交互注意力模块和正则化项，限制模型的复杂度，增强表达特征能力，提高泛化性能，进而提升重识别能力。将改进模型在Market1501、DukeMTMC数据集上进行验证，实验结果表明，相较于原始模型，平均精度均值在Market1501数据集上提升了5.6个百分点；在DukeMTMC数据集上提升了5.9个百分点，表明了改进后算法的有效性。
Abstract：The interference factors such as the diversity of pedestrian pose variations and the complexity of background environments can lead to inadequate capability to express pedestrian features, thereby affecting the accuracy of pedestrian re-identification.This paper aims to improve the AlignedReID++ model based on deep learning by integrating the DenseNet121 network with the backbone ResNet50 network of AlignedReID++.Leveraging the strengths of both networks enhances the model's feature extraction capability.By introducing a cross-dimensional interactive attention module and regularization terms, the model's complexity is constrained, its feature representation capability is strengthened, and its generalization performance is improved, thereby boosting re-identification accuracy.Experimental validation on the Market1501 and DukeMTMC datasets demonstrates that, compared to the original model, the improved model achieves increases by 5.6 percentage points in MAP on the Market1501 dataset, and 5.9 percentage points in MAP on the DukeMTMC dataset.These results confirm the effectiveness of the proposed algorithm.
[1] 胡海峰，倪宗煜，赵海涛，等.无人机场景下基于Transformer的轻量化行人重识别[J].南京邮电大学学报(自然科学版),2024,44(3):48-62.HU H F,NI Z Y,ZHAO H T,et al.Transformer based lightweight person re-identification in unmanned aerial vehicle images[J].Journal of Nanjing University of Posts and Telecommunications(Natural Science Edition),2024,44(3):48-62.(in Chinese)
[2] 李倩.基于自监督部位感知的行人重识别模型及其在铁路客运站的应用[J].铁路计算机应用，2024,33(2):19-23.LI Q.Pedestrian reidentification model based on self-supervised part perception and its application in railway passenger stations[J].Railway Computer Application,2024,33(2):19-23.(in Chinese)
[3] 蔺丽华，燕梦梦，廖晓群.行人重识别技术在洗煤厂中的应用研究[J].煤炭技术，2024,43(2):205-208.LIN L H,YAN M M,LIAO X Q.Research on application of person re-identification in coal washing plant[J].Coal Technology,2024,43(2):205-208.(in Chinese)
[4] 赵安新，杨金桥，杨浩波，等.基于改进DeepSORT和FastReID的室内多目标人员跨镜识别与跟踪[J].西安科技大学学报，2023,43(3):622-630.ZHAO A X,YANG J Q,YANG H B,et al.Indoor multi-object personnel recognition and tracking across camera based on optimized DeepSORT and FastReID[J].Journal of Xi’an University of Science and Technology,2023,43(3):622-630.(in Chinese)
[5] 金梅，任婷婷，张立国，等.改进的轻量级行人目标检测算法[J].计量学报，2024,45(2):186-193.JIN M,REN T T,ZHANG L G,et al.Improved lightweight pedestrian target detection algorithm[J].Acta Metrologica Sinica,2024,45(2):186-193.(in Chinese)
[6] 孔凡国，王鑫，仇展明，等.基于改进YOLOv7的复杂行人检测系统研究[J].机械工程与自动化，2024(6):58-61.KONG F G,WANG X,QIU Z M,et al.Research on a complex pedestrian detection system based on improved YOLOv7[J].Mechanical Engineering & Automation,2024(6):58-61.(in Chinese)
[7] YE M,SHEN J B,LIN G J,et al.Deep learning for person re-identification:a survey and outlook[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2022,44(6):2872-2893.
[8] 张立亚，王寓，郝博南.基于改进度量学习的煤矿井下行人重识别方法研究[J].工矿自动化，2023,49(9):84-89,166.ZHANG L Y,WANG Y,HAO B N.Research on personnel re-recognition method in coal mine underground based on improved metric learning[J].Journal of Mine Automation,2023,49(9):84-89,166.(in Chinese)
[9] 黄海新，陶文博，杜亭亭.基于度量学习的行人重识别综述[J].沈阳理工大学学报，2023,42(5):1-10,17.HUANG H X,TAO W B,DU T T.Review of pedestrian re-identification based on metric learning[J].Journal of Shenyang Ligong University,2023,42(5):1-10,17.(in Chinese)
[10] TANG Q W,YAN P,CHEN J,et al.Person re-identification based on multi-scale global feature and weight-driven part feature[J].AI Communications,2022,35(3):207-223.
[11] 焦传扬，丁学明.基于全局滤波池化多关系Transformer网络的行人重识别[J].控制工程，2024,31(5):912-919.JIAO C Y,DING X M.Person re-identification based on global filter pooling multi relation transformer network[J].Control Engineering of China,2024,31(5):912-919.(in Chinese)
[12] YU M X,WANG J,YOU R,et al.Multiple-local feature and attention fused person re-identification method[J].Intelligent Data Analysis,2024,28(6):1679-1695.
[13] 王琦，刘志刚，王淼，等.姿态驱动的局部特征对齐的行人重识别[J].计算机系统应用，2023,32(4):268-273.WANG Q,LIU Z G,WANG M,et al.Pose-driven person re-identification with local feature alignment[J].Computer Systems and Applications,2023,32(4):268-273.(in Chinese)
[14] YIN K N,DING Z,DONG Z H,et al.Person re-identification method based on fine-grained feature fusion and self-attention mechanism[J].Computing,2024,106(5):1681-1705.
[15] 吴宇森，于宝华，荣江，等.基于多尺度注意力与特征融合的行人重识别方法研究[J].石河子大学学报(自然科学版),2025,43(1):122-132.WU Y S,YU B H,RONG J,et al.Research on pedestrian re-identification method based on multi-scale attention and feature fusion[J].Journal of Shihezi University(Natural Science),2025,43(1):122-132.(in Chinese)
[16] LUO H,JIANG W,ZHANG X,et al.AlignedReID++:dynamically matching local information for person re-identification[J].Pattern Recognition,2019,94:53-61.
[17] 杨世欣，胡晓光，杜卓群，等.基于跨纬度交互注意力机制的行人重识别方法[J].智能计算机与应用，2022,12(3):28-32,39.YANG S X,HU X G,DU Z Q,et al.Research on person re-identification based on triplet attention mechanism[J].Intelligent Computer and Applications,2022,12(3):28-32,39.(in Chinese)
[18] 张子昊，周千里，王蓉.基于空间注意力机制的行人再识别方法[J].北京航空航天大学学报，2020,46(9):1747-1755.ZHANG Z H,ZHOU Q L,WANG R.Pedestrian re-identification method based on spatial attention mechanism[J].Journal of Beijing University of Aeronautics and Astronautics,2020,46(9):1747-1755.(in Chinese)
[19] 谢凌望，陈满，彭宇瑞，等.基于ResNet和DenseNet模型的铝罐表面涂层缺陷检测方法[J].汕头大学学报(自然科学版),2024,39(4):40-48.XIE L W,CHEN M,PENG Y R,et al.A surface coating defect detection method for aluminum cans based on resnet and densenet models[J].Journal of Shantou University(Natural Science Edition),2024,39(4):40-48.(in Chinese)
[20] LIANG Z,SHEN L Y,LU T,et al.Person re-identification meets image search[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Los Alamitos,USA:IEEE,2015:4321-4330.
[21] RISTANI E,SOLERA F,ZOU R S,et al.Performance measures and a data set for multi-target,multi-camera tracking[C]//Computer Vision-ECCV 2016 Workshops.Amsterdem,The Netherlands:Springer,2016:17-35.
[22] HU J,SHEN L,SUN G,et al.Squeeze-and-excitation networks[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2020,42(8):2011-2023.
[23] 耿韶松，李晋国.基于动态卷积与注意力的多特征融合行人重识别[J].计算机工程与设计，2023,44(4):1228-1234.GENG S S,LI J G.Person re-identification based on multi-feature fusion of dynamic convolution and attention[J].Computer Engineering and Design,2023,44(4):1228-1234.
[24] ZHONG C Y,QI G Q,MAZUR N,et al.A domain adaptive person re-identification based on dual attention mechanism and camstyle transfer[J].Algorithms,2021,14(12):361-375.
[25] SARFRAZ M S,SCHUMANN A,EBERLE A,et al.A pose-sensitive embedding for person re-identification with expanded cross neighborhood re-ranking[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.Salt Lake City,UT,USA:IEEE,2018:420-429.
[26] LI R,ZHANG B P,ZHU T,et al.A divide-and-unite deep network for person re-identification[J].Applied Intelligence,2020,51(3):1-13.
[27] 宋晓茹，杨佳，高嵩，等.基于注意力机制与多尺度特征融合的行人重识别方法[J].科学技术与工程，2022,22(4):1526-1533.SONG X R,YANG J,GAO S,et al.Person re-identification method based on attention mechanism and multi-scale feature fusion[J].Science Technology and Engineering,2022,22(4):1526-1533.(in Chinese)
基本信息:
中图分类号:TP391.41;TP18
引用信息:
[1]宋建辉,马赫遥,赵亚威.基于改进AlignedReID++的行人重识别方法[J].沈阳理工大学学报,2026,45(02):1-8+17.
基金信息:
辽宁省属本科高校基本科研业务费专项资金资助项目(LJ212410144053)
2026-02-24
2026-02-24