Vehicle Re-identification Algorithm Based on Component Fusion Feature

doi:10.19678/j.issn.1000-3428.0052284

Computer Engineering ›› 2019, Vol. 45 ›› Issue (6): 12-20. doi: 10.19678/j.issn.1000-3428.0052284

Special Issue: ；

Previous Articles Next Articles

Vehicle Re-identification Algorithm Based on Component Fusion Feature

Xiying LI^1,^2,^3,⁴,Zhihao ZHOU^1,^2,^3,⁴,Mingkai QIU^1,^2,^3,⁴

^1. School of Intelligent Systems Engineering,Sun Yat-sen University,Guangzhou 510006,China
^2. Guangdong Province Key Laboratory of Intelligent Transportation System,Guangzhou 510006,China
^3. Key Laboratory of Video and Image Intelligent Analysis and Application Technology,Ministry of Public Security of PRC,Guangzhou 510006,China
^4. National Engineering Laboratory of Video and Image Information Intelligent Analysis and Sharing Application Technology,Beijing 100048,China

Received:2018-08-03 Revised:2018-09-05 Online:2019-06-15 Published:2019-06-17

基于部件融合特征的车辆重识别算法

李熙莹^1,^2,^3,⁴,周智豪^1,^2,^3,⁴,邱铭凯^1,^2,^3,⁴

^1. 中山大学智能工程学院,广州 510006
^2. 广东省智能交通系统重点实验室,广州 510006
^3. 视频图像智能分析与应用技术公安部重点实验室,广州 510006
^4. 视频图像信息智能分析与共享应用技术国家工程实验室,北京 100048

作者简介:李熙莹(1972—),女,副教授、博士,主研方向为交通信息技术、视频图像、视频大数据技术|周智豪、邱铭凯,硕士研究生。
基金资助:
国家自然科学基金“视频大数据高效表达、深度分析与综合利用”(U1611461)

Abstract

Abstract:

To address the re-identification problem of different individual vehicles with identical types,a new vehicle re-identification algorithm is proposed.According to the component detection algorithm,the window and the vehicle face region with large differences between different vehicles are obtained,and the vehicle features of the detected vehicle window and the vehicle face region are extracted and merged to generate new fusion features.The distance measurement between image features is calculated for classification and recognition.The test is carried out on the public dataset VRID-1 of Sun Yat-sen university and results show that the Rank1 matching rate of the algorithm reaches 66.67%,which is obviously better than the classical traditional feature representation algorithm,thus verifies the feasibility and validity of the algorithm.

Key words: vehicle re-identification, component detection, feature extraction, feature fusion, distance measurement

摘要：

针对车辆型号相同但车辆个体不同的重识别问题,提出一种新的车辆重识别算法。运用部件检测算法获取不同车辆之间差异较大的车窗和车脸区域,对检测到的车窗和车脸区域进行特征提取并进行融合,生成新的融合特征,计算图像特征之间距离度量进行分类识别。在中山大学公开数据集VRID-1上进行测试,结果表明,该算法的Rank1匹配率达到66.67%,明显优于经典的传统特征表征算法,从而验证该算法是可行且有效的。

关键词: 车辆重识别, 部件检测, 特征提取, 特征融合, 距离度量

Xiying LI, Zhihao ZHOU, Mingkai QIU. Vehicle Re-identification Algorithm Based on Component Fusion Feature[J]. Computer Engineering, 2019, 45(6): 12-20.

李熙莹, 周智豪, 邱铭凯. 基于部件融合特征的车辆重识别算法[J]. 计算机工程, 2019, 45(6): 12-20.

/ Recommend / Download Citations

URL: https://www.ecice06.com/EN/10.19678/j.issn.1000-3428.0052284

https://www.ecice06.com/EN/Y2019/V45/I6/12

Figures/Tables 18

References 15

[1]	MEI L T, ,LAM W H K . Application of automatic vehicle identification technology for real-time journey time estimation[J]. Information Fusion, 2011,12(1):11-19. doi: 10.1016/j.inffus.2010.01.002
[2]	张耿宁, 王家宝, 张亚非 , 等. 基于特征融合的行人重识别方法[J]. 计算机工程与应用, 2017,53(12):185-189.
[3]	刘娜 . 基于卷积神经网络的行人重识别算法[D]. 上海:华东师范大学, 2017.
[4]	CHENG Deng, 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.Las Vegas, USA: IEEE Press, 2016: 1335-1344.
[5]	TIAN Yin, DONG Honghui, JIA Limin , et al. A vehicle re-identification algorithm based on multi-sensor correlation[J]. Journal of Zhejiang University-SCIENCE C (Computers and Electronics), 2014,15(5):372-382.
[6]	COIFMAN B. Vehicle reidentification and travel time measurement,part II:uncongested freeways and the onset of congestion[C]//Proceedings of 2001 IEEE Intelligent Transportation Systems.Oakland, USA: IEEE Press, 1999: 899-917.
[7]	王盼盼, 李玉惠 . 基于特征融合和L-M算法的车辆重识别方法[J]. 电子科技, 2018,4(1):12-15. URL
[8]	ZHANG Yiheng, LIU Dong, ZHA Zhengjun. Improving triplet-wise training of convolutional neural network for vehicle re-identification[C]//Proceedings of IEEE International Conference on Multimedia and Expo.Washington D.C., USA: IEEE Press, 2017: 1386-1391.
[9]	XU Qingtong, YAN Ke, TIAN Yonghong. Learning a repression network for precise vehicle search[C]//Proceedings of IEEE CVPR’17.Washington D.C., USA: IEEE Press, 2017: 125-136.
[10]	ZAPLETAL D, HEROUT A. Vehicle re-identification for automatic video traffic surveillance [C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition Workshops.[S.1.]:IEEE Computer Society, 2016: 1568-1574.
[11]	ZHOU Yi, LIU Li, SHAO Ling . Vehicle re-identification by deep hidden multi-view inference[J]. IEEE Transactions on Image Processing, 2018,27(7):3275-3287.
[12]	REN Shaoqing, GIRSHICK R . Faster R-CNN:towards real-time object detection with region proposal networks[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017,39(6):1137-1149.
[13]	SZEGEDY C, LIU Wei, JIA Yangqing , et al. Going deeper with convolutions[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C., USA: IEEE Press, 2015: 1-9.
[14]	MACQUEEN J. Some methods for classification and analysis of multi-variate observations[C]//Proceedings of Berkeley Symposium on Mathematical Statistics and Probability.Berkeley, USA: California University Press, 1967: 281-297.
[15]	LI Xiying, YUAN Minxian, JIANG Qianyin , et al. VRID-1:a basic vehicle re-identification dataset for similar vehicles[C]//Proceedings of IEEE International Conference on Intelligent Transportation Systems.Washington D.C.,USA:IEEE Press, 2017: 1-8.

方法	Rank 1	Rank 5	Rank 10	Rank 15
实验1	1.67	4.00	5.67	6.00
实验2	48.67	75.00	83.67	88.67
本文算法	66.67	82.67	90.00	92.33

方法	Rank 1	Rank 5	Rank 10	Rank 15
实验1	1.67	4.00	5.67	6.00
实验2	48.67	75.00	83.67	88.67
本文算法	66.67	82.67	90.00	92.33

方法	Rank 1	Rank 10	Rank 20	Rank 30	Rank 40	Rank 50
实验1	1.37	2.26	3.93	5.19	6.22	7.44
实验2	2.35	34.04	46.19	53.56	59.22	63.67
本文算法	3.11	43.26	50.67	62.81	68.11	72.26

方法	Rank 1	Rank 10	Rank 20	Rank 30	Rank 40	Rank 50
实验1	1.37	2.26	3.93	5.19	6.22	7.44
实验2	2.35	34.04	46.19	53.56	59.22	63.67
本文算法	3.11	43.26	50.67	62.81	68.11	72.26

Please choose a citation manager

Content to export