A simple feature extraction method for human shape detection

被引：0

作者：

Wang, SM ^{[1
]}

Liu, NQ ^{[1
]}

机构：

[1] Univ Elect Sci & Technol China, Sch Comp Sci & Engn, Chengdu, Peoples R China

来源：

APPLICATIONS OF DIGITAL TECHNIQUES IN INDUSTRIAL DESIGN ENGINEERING-CAID&CD' 2005 | 2005年

关键词：

computer vision; image understanding; human detection; features extraction; motion analysis; neural network;

D O I：

暂无

中图分类号：

TP39 [计算机的应用];

学科分类号：

081203 ; 0835 ;

摘要：

A novel feature extraction method for detecting human shapes from video images is proposed in this paper. First boundary contours of moving objects are extracted from one of consecutive video frames in this system. Then 3 matching parameters are calculated. Lastly likelihood values and matching parameters are fed to the back- propagation neural network module to judge whether moving object is human or not and analyze moving direction and postures.

引用

页码：574 / 577

页数：4

共 7 条

[1]

CANNY JF, IEEE T PATTERN ANAL, V8, P901

[2]

Collins R., 2000, VSAM FINAL REP, V5, P1

[3]

Karmann K., 1990, Moving object recognition using an adaptive background memory'

[4]

Kuno Y., 1996, Proceedings of the 13th International Conference on Pattern Recognition, P865, DOI 10.1109/ICPR.1996.547291

[5] Moving target classification and tracking from real-time video [J].

Lipton, AJ ;

Fujiyoshi, H ;

Patil, RS .

FOURTH IEEE WORKSHOP ON APPLICATIONS OF COMPUTER VISION - WACV'98, PROCEEDINGS, 1998, :8-14

[6]

MAGGIONI C, 1998, GESTURE COMPUTER HIS

[7]

TATE S, 2001, VIDEO BASED HUMAN SH, V20, P1090

← 1 →