Vehicle Tracking in Video Based on Pixel Level Motion Vector

被引：0

作者：

Xiong, Yang ^{[1
]}

Lu, Xiaobo ^{[1
]}

Zhu, Zhou ^{[2
]}

Zeng, Weili ^{[2
]}

机构：

[1] Southeast Univ, Sch Automat, Nanjing 210096, Jiangsu, Peoples R China

[2] Southeast Univ, Sch Transportat, Nanjing 210096, Peoples R China

来源：

MULTIMEDIA AND SIGNAL PROCESSING | 2012年 / 346卷

基金：

中国国家自然科学基金;

关键词：

vehicle tracking; vehicle detection; block matching; motion vector;

D O I：

暂无

中图分类号：

TP301 [理论、方法];

学科分类号：

081202 ;

摘要：

In this paper, the problem of missing vehicles halfway in previous approach of vehicle tracking based on motion vector is studied, and a vehicle tracking algorithm based on pixel level motion vector is proposed. In the proposed algorithm, blocks of vehicles are shifted by pixel level motion vector which is acquired directly by block matching method, and overlapping between blocks contained in a single vehicle is allowed. By the experiments, the proposed algorithm was proved to be very successful. It can track vehicles farther than block level motion vector based approach.

引用

页码：200 / +

页数：2

共 6 条

[1]

Ha D.B., 2010, COMPUTER ENG, V36, P197

[2]

Kamijo S, 2000, INT C PATT RECOG, P140, DOI 10.1109/ICPR.2000.905292

[3] A traffic accident recording and reporting model at intersections [J].

Ki, Yong-Kul ;

Lee, Dong-Young .

IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS, 2007, 8 (02) :188-194

[4] Optical flow based vehicle tracking strengthened by statistical decisions [J].

Nejadasl, Fatemeh Karimi ;

Gorte, Ben G. H. ;

Hoogendoorn, Serge P. .

ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING, 2006, 61 (3-4) :159-169

[5]

Zhaoxiang Zhang, 2010, Proceedings of the 2010 20th International Conference on Pattern Recognition (ICPR 2010), P1771, DOI 10.1109/ICPR.2010.437

[6]

Zhu Z., 2007, J TRANSPORTATION ENG, V5, P110

← 1 →