Abstract—From past few years, Gait based recognition is one of the emerging new biometric identification technology for human identification, surveillance and other security applications. In some previous research articles GEI has been reported as a good feature robust to silhouette errors and image noise, but it ignores some gait motion information. So, this paper proposes a novel algorithm which uses Energy Deviation Image (EDI) and based on Fuzzy Principal Component Analysis(FPCA). Firstly, original gait sequences are preprocessed and Enhanced Energy Deviation Image is obtained. Secondly, using Fuzzy Principal Component Analysis eigen values and eigenvectors are formulated and which are finally termed as Fuzzy Components. Then, dimensional space is reduced by projecting eigenvectors into low dimensional space. At last, NN classifier is utilized in feature classification. The algorithm is tested on different datasets on CASIA database and the experimental results show that this algorithm achieves higher recognition performance.
Index Terms—Energy deviation image, fuzzy principal component analysis, gait period, feature extraction, gait recognition.
The authors are with the Computer Science and Engineering Department,H.B.T.I., Kanpur, India (e-mail: email@example.com,firstname.lastname@example.org).
Cite: Rohit Katiyar, Member, IACSIT and Vinay Kumar Pathak, "Gait Recognition Based on Energy Deviation Image Using Fuzzy Component Analysis," International Journal of Innovation, Management and Technology vol. 4, no. 1, pp. 43-46, 2013.