内容一致性行人重识别算法

doi:10.19678/j.issn.1000-3428.0056743

计算机工程 ›› 2021, Vol. 47 ›› Issue (3): 237-242. doi: 10.19678/j.issn.1000-3428.0056743

内容一致性行人重识别算法

田智慧^1,2, 郑付科¹, 高需³

1. 郑州大学信息工程学院, 郑州 450001;
2. 郑州大学地球科学与技术学院, 郑州 450052;
3. 郑州大学河南省超级计算中心, 郑州 450052

收稿日期:2019-11-28 修回日期:2020-01-16 发布日期:2020-03-05
作者简介:田智慧(1965-),男,教授、博士,主研方向为智慧城市、深度学习;郑付科,硕士研究生;高需,讲师、博士。
基金资助:
国家重点研发计划（2018YFB0505004-03）；郑州大学科研启动基金（32210919）。

Content-Consistent Pedestrian Re-identification Algorithm

TIAN Zhihui^1,2, ZHENG Fuke¹, GAO Xu³

1. School of Information Engineering, Zhengzhou University, Zhengzhou 450001, China;
2. School of Geoscience and Technology, Zhengzhou University, Zhengzhou 450052, China;
3. Supercomputing Center of Henan Province, Zhengzhou University, Zhengzhou 450052, China

Received:2019-11-28 Revised:2020-01-16 Published:2020-03-05

摘要/Abstract

摘要： 行人重识别是指利用计算机视觉技术识别不同监控设备下的目标行人，该技术在公共安全与相册管理等方面应用较广。然而现有行人重识别算法在局部特征区域划分后出现离异值使该区域内容不一致，导致局部特征可区分性降低。提出一种基于局部区域特征选择的内容一致性行人重识别算法。将行人图像输入残差卷积神经网络取得张量，根据局部区域内容一致性从张量中选择基本单位特征向量，使用Softmax函数计算其局部区域概率重新生成局部区域，从而消除离异值，增加类间差异并减少类内差异。实验结果表明，与Spindel、PN-GAN等行人重识别算法相比，该算法的行人重识别准确率更高，其提取的行人特征可区分性和鲁棒性更好。

关键词: 行人重识别, 公共安全, 内容一致性, 局部特征, 离异值

Abstract: Pedestrian re-identification is to use computer vision technology to identify the target pedestrian under different monitoring equipment,which is widely used in public safety and album management.However,the existing pedestrian re-identification algorithms often generate outlier values after the local feature region is divided and thus lead to the region content inconsistency, reducing the distinguishability of local features.This paper proposes a Content-Consistent Pedestrian Re-identification(CCreID)algorithm based on local region feature selection.The pedestrian image is input into the residual convolutional neural network to obtain the tensor.According to the consistency of local region content,the basic unit feature vector is selected from the tensor.Then the local region probability is calculated by using the Softmax function to regenerate the local region,so as to eliminate the outliers,increase the differences between classes and reduce the differences within classes.Experimental results show that compared with Spindel,PN-GAN and other pedestrian re-identification algorithms,the proposed algorithm has higher accuracy of pedestrian re-identification and better distinguishability and robustness of extracted pedestrian features.

Key words: pedestrian re-identification, public security, content consistency, local feature, outlier

中图分类号:

TP391.41

田智慧, 郑付科, 高需. 内容一致性行人重识别算法[J]. 计算机工程, 2021, 47(3): 237-242.

TIAN Zhihui, ZHENG Fuke, GAO Xu. Content-Consistent Pedestrian Re-identification Algorithm[J]. Computer Engineering, 2021, 47(3): 237-242.

http://www.ecice06.com/CN/Y2021/V47/I3/237

图/表 13

20210322172255

20210322172915

20210322172919

20210322172922

20210322172925

20210322172929

20210322172931

20210322172934

20210322172938

20210322172942

20210322172945

20210322172949

20210322172954

参考文献

[1] DING Zongyuan,WANG Hongyuan,CHEN Fuhua,et al.Person re-identification based on distance centralization and projection vectors learning[J].Journal of Computer Research and Development,2017,54(8):1785-1794.(in Chinese)丁宗元,王洪元,陈付华,等.基于距离中心化与投影向量学习的行人重识别[J].计算机研究与发展,2017,54(8):1785-1794.
[2] SONG Wanru,ZHAO Qingqing,CHEN Changhong,et al.Survey on pedestrian re-identification research[J].CAAI Transactions on Intelligent Systems,2017,12(6):770-780.(in Chinese)宋婉茹,赵晴晴,陈昌红等.行人重识别研究综述[J].智能系统学报,2017,12(6):770-780.
[3] ZHENG L,YANG Y,HAUPTMANN A G,et al.Person re-identification:past,present and future[EB/OL].[2019-10-13].https://www.researchgate.net/publication/308981025_Person_Re-identification_Past_Present_and_Future.
[4] LIU Xiaokai,WANG Hongyu,WU Yi,et al.An ensemble color model for human re-identification[C]//Proceedings of 2015 IEEE Winter Conference on Applications of Computer Vision.Washington D.C.,USA:IEEE Press,2015:868-875.
[5] FARENZENA M,BAZZANI L,PERINA A,et al.Person re-identification by symmetry-driven accumulation of local features[C]//Proceedings of 2010 IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2010:2360-2367.
[6] HAMDOUN O,MOUTARDE F,STANCIULESCU B,et al.Person re-identification in multi-camera system by signature based on interest point descriptors collected on short video sequences[C]//Proceedings of 2008 IEEE Conference on Distributed Smart Cameras.Washington D.C.,USA:IEEE Press,2008:1-6.
[7] LI Zhi,SUN Yubao,WANG Feng,et al.Clothing image classification and retrieval algorithm based on deep convolutional neural network[J].Computer Engineering,2016,42(11):309-315.(in Chinese)厉智,孙玉宝,王枫,等.基于深度卷积神经网络的服装图像分类检索算法[J].计算机工程,2016,42(11):309-315.
[8] CHENG De,GONG Yihong,ZHOU Sanping,et al.Person re-identification by multi-channel parts-based CNN with improved triplet loss function[C]//Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2016:1335-1344.
[9] CAO Z,SIMON T,WEI S E,et al.Realtime multi-person 2D pose estimation using part affinity fields[C]//Proceedings of 2017 IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2017:7291-7299.
[10] ZHAO Haiyu,TIAN Maoqing,SUN Shuyang,et al.Spindle net:person re-identification with human body region guided feature decomposition and fusion[C]//Proceedings of 2017 IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2017:1077-1085.
[11] YAO Hantao,ZHANG Shiliang,HONG Richang,et al.Deep representation learning with part loss for person re-identification[J].IEEE Transactions on Image Processing,2019,28(6):2860-2871.
[12] LIU Xihui,ZHAO Haiyu,TIAN Maoqing,et al.Hydraplus-net:attentive deep features for pedestrian analysis[C]//Proceedings of 2017 IEEE Conference on Computer Vision.Washington D.C.,USA:IEEE Press,2017:350-359.
[13] ZHAO Liming,LI Xi,WANG Jingdong,et al.Deeply-learned part-aligned representations for person re-identification[C]//Proceedings of 2017 IEEE Conference on Computer Vision.Washington D.C.,USA:IEEE Press,2017:3219-3228.
[14] SUN Yifan,ZHENG Liang,YANG Yi,et al.Beyond part models:person retrieval with refined part pooling (and a strong convolutional baseline)[C]//Proceedings of 2018 European Conference on Computer Vision.Washington D.C.,USA:IEEE Press,2018:480-496.
[15] ZHENG Liang,SHEN Liyue,TIAN Lu,et al.Scalable person re-identification:a benchmark[C]//Proceedings of 2015 IEEE International Conference on Computer Vision.Washington D.C.,USA:IEEE Press,2015:1116-1124.
[16] RISTANI E,SOLERA F,ZOU R,et al.Performance measures and a data set for multi-target,multi-camera tracking[C]//Proceedings of 2016 European Conference on Computer Vision.Berlin,Germany:Springer,2016:17-35.
[17] HE Kaiming,ZHANG Xiangyu,REN Shaoqing,et al.Deep residual learning for image recognition[C]//Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2016:770-778.
[18] VARIOR R R,SHUAI B,LU J,et al.A siamese long short-term memory architecture for human re-identification[C]//Proceedings of 2016 European Conference on Computer Vision.Berlin,Germany:Springer,2016:135-153.
[19] YAO Hantao,ZHANG Shiliang,HONG Richang,et al.Deep representation learning with part loss for person re-identification[J].IEEE Transactions on Image Processing,2019,28(6):2860-2871.
[20] FELZENSQWALB P F,MCALLESTER D,RAMANAN D,et al.A discriminatively trained,multiscale,deformable part model[C]//Proceedings of 2008 IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2008:1-8.
[21] QIAN Xuelin,FU Yanwei,XIANG Tao,et al.Pose-normalized image generation for person re-identification[C]//Proceedings of 2018 European Conference on Computer Vision.Berlin,Germany:Springer,2018:650-667.

选择文件类型/文献管理软件名称

选择包含的内容

内容一致性行人重识别算法

Content-Consistent Pedestrian Re-identification Algorithm

RichHTML

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 13

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	韩华, 黄丽, 田瑾, 王春媛. 基于双中间模态的四流网络跨模态行人重识别[J]. 计算机工程, 2023, 49(8): 302-309.
[2]	惠子薇, 何坤, 冯犇, 苏曜. 基于视觉特性的图像质量评价[J]. 计算机工程, 2023, 49(7): 189-195.
[3]	沈秀轩, 吴春雷, 冯叶棋, 程铭, 张俊三, 朱杰. 基于双分支特征融合的医学报告生成方法[J]. 计算机工程, 2023, 49(6): 274-283,291.
[4]	魏紫薇, 屈丹, 柳聪. 基于连接注意力的行人重识别特征提取方法[J]. 计算机工程, 2022, 48(7): 220-226.
[5]	郝阿香, 贾郭军. 结合注意力与批特征擦除的行人重识别模型[J]. 计算机工程, 2022, 48(7): 270-276,306.
[6]	刘晓蓉, 李小霞, 秦昌辉. 融合多尺度对比池化特征的行人重识别方法[J]. 计算机工程, 2022, 48(4): 292-298.
[7]	符进武, 范自柱, 石林瑞, 郭心悦, 黄祎婧. 基于多尺度多粒度融合的行人重识别方法[J]. 计算机工程, 2022, 48(3): 271-279.
[8]	汪荣贵, 李懂, 杨娟, 薛丽霞. 基于跨域特征关联与聚类的无监督行人重识别[J]. 计算机工程, 2022, 48(3): 229-235,243.
[9]	曾涛, 薛峰, 杨添. 面向行人重识别的通道与空间双重注意力网络[J]. 计算机工程, 2022, 48(12): 281-287,295.
[10]	徐智明, 戚湧. 基于UV贴图优化人体特征的行人重识别[J]. 计算机工程, 2022, 48(11): 83-88,95.
[11]	王福银, 韩华, 黄丽, 陈益平. 时间特征互补的无监督视频行人重识别[J]. 计算机工程, 2022, 48(10): 313-320.
[12]	李佳宾, 李学伟, 刘宏哲, 徐成. 基于局部特征关联与全局注意力机制的行人重识别[J]. 计算机工程, 2022, 48(1): 245-252.
[13]	董亚超, 刘宏哲, 徐成. 基于显著性多尺度特征协作融合的行人重识别方法[J]. 计算机工程, 2021, 47(6): 234-244,252.
[14]	胡晓强, 魏丹, 王子阳, 沈江霖, 任洪娟. 基于时空关注区域的视频行人重识别[J]. 计算机工程, 2021, 47(6): 277-283.
[15]	徐龙壮, 彭力, 朱凤增. 多任务金字塔重叠匹配的行人重识别方法[J]. 计算机工程, 2021, 47(1): 239-245,254.

模态框（Modal）标题

选择文件类型/文献管理软件名称

选择包含的内容

内容一致性行人重识别算法

Content-Consistent Pedestrian Re-identification Algorithm

RichHTML

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 13

参考文献

相关文章 15

编辑推荐

Metrics

本文评价