Tracking human motion in a complex scene using textural analysis

Author

Nugroho, Hertog ; Hwang, Junghyun ; Ozawa, Shinji

Author_Institution

Dept. of Electr. Eng., Keio Univ., Yokohama, Japan

Volume

2

fYear

1994

fDate

5-9 Sep 1994

Firstpage

727

Abstract

A texture-based system which is able to track human motion is proposed. The system consists of four procedures: motion extraction, object´s definition, textural feature extraction, and object identification. Motion extraction is done by temporal differentiation method while Hausdorff distance method is used to define objects´ existence. Texture characteristics are calculated in a polar coordinate domain, considering the nature of the problem. Identification is done in an arbitrary condition, including occluded objects and the result is discussed

Keywords

computer vision; feature extraction; image texture; motion estimation; object recognition; tracking; Hausdorff distance method; complex scene; human motion tracking; motion extraction; object definition; object identification; occluded objects; polar coordinate domain; temporal differentiation; textural analysis; textural feature extraction; Clothing; Ergonomics; Feature extraction; Humans; Layout; Lighting; Medical treatment; Motion analysis; Shape; Tracking;

fLanguage

English

Publisher

ieee

Conference_Titel

Industrial Electronics, Control and Instrumentation, 1994. IECON '94., 20th International Conference on

Conference_Location

Bologna

Print_ISBN

0-7803-1328-3

Type

conf

DOI

10.1109/IECON.1994.397875

Filename

397875