A Human Action Recognition Method Based on Tchebichef Moment Invariants and Temporal Templates

Author

Lu, Yanan ; Li, Yakang ; Shen, Yang ; Ding, Fang ; Wang, Xiaofeng ; Hu, Jicheng ; Ding, Songtao

Author_Institution

State Key Lab. of Software Eng., Wuhan Univ., Wuhan, China

Volume

2

fYear

2012

fDate

26-27 Aug. 2012

Firstpage

76

Lastpage

79

Abstract

In this paper, a new human action recognition method based on Tchebichef moment invariants and temporal templates is presented. We use the motion energy image (MEI) and motion history image (MHI) as the feature representation of the human action at first. Then the Tchebichef moment invariants extract the feature vectors of MEI and MHI. Tchebichef moment invariants perform better than Hu moment invariants and Zernike moment invariants. Finally cluster the actions and use the nearest neighbor algorithm to recognize each human action. The result of these experiments suggests that this method has a high recognition rate in in both noise-free and noisy condition. Therefore, the algorithm has a good robustness.

Keywords

feature extraction; image motion analysis; pattern classification; vectors; Hu moment invariants; MEI; MHI; Tchebichef moment invariants; Zernike moment invariants; feature vectors extraction; human action recognition; motion energy image; motion history image; nearest neighbor algorithm; temporal templates; Educational institutions; Feature extraction; Humans; Image recognition; Manganese; Noise; Vectors; MEI; MHI; Tchebichef moment invariants; cluster; human action recognition;

fLanguage

English

Publisher

ieee

Conference_Titel

Intelligent Human-Machine Systems and Cybernetics (IHMSC), 2012 4th International Conference on

Conference_Location

Nanchang, Jiangxi

Print_ISBN

978-1-4673-1902-7

Type

conf

DOI

10.1109/IHMSC.2012.114

Filename

6305728