Gait recognition based on lower limb

Author

Yaacob, Nurul Iliani ; Tahir, Nooritawati Md ; Abdullah, R.

Author_Institution

Fac. of Electr. Eng., Univ. Teknol. MARA (UiTM), Shah Alam, Malaysia

fYear

2012

fDate

16-17 July 2012

Firstpage

294

Lastpage

297

Abstract

In this paper, a new approach is proposed for extracting feature from walking human based on the lower body. This approach utilized DCT to extract features followed by PCA as feature selection. Further ANN and PNN are utilised as classifiers. All developed method is tested with 10 subjects from CASIA gait database. Initial findings showed that the proposed method is viable based on 90% recognition rate for ANN and 80% for PNN.

Keywords

discrete cosine transforms; feature extraction; feedforward neural nets; gait analysis; image recognition; principal component analysis; CASIA gait database; DCT; PCA; PNN; discrete cosine transform; feature extraction; feature selection; gait recognition; lower limb; principal component analysis; probabilistic neural network; Artificial neural networks; Discrete cosine transforms; Feature extraction; Humans; Legged locomotion; Principal component analysis; artificial neural network; averaged silhouette; discrete cosine transform; gait recognition; principal component analysis; probabilistic neural network;

fLanguage

English

Publisher

ieee

Conference_Titel

Control and System Graduate Research Colloquium (ICSGRC), 2012 IEEE

Conference_Location

Shah Alam, Selangor

Print_ISBN

978-1-4673-2035-1

Type

conf

DOI

10.1109/ICSGRC.2012.6287179

Filename

6287179