International Journal of Computer Applications |
Foundation of Computer Science (FCS), NY, USA |
Volume 117 - Number 17 |
Year of Publication: 2015 |
Authors: Riant Kaur, Heena |
10.5120/20643-3350 |
Riant Kaur, Heena . An Approach Defining Gait Recognition System using K-Means and MDA. International Journal of Computer Applications. 117, 17 ( May 2015), 1-4. DOI=10.5120/20643-3350
Biometric Authentication is a technology of verifying and identifying individuals. Gait Recognition is a type of biometric identification method which is related to behavioral characteristic of a person. Gait identifies a person based on the manner he moves or walks on foot. Gait needs reduced detail and is difficult to conceal Gait can identify people from a distance without their cooperation. Gait can be used in real time environments to detect threats and frauds. Gait can be captured using moving video, floor sensors or motion recording sensors. In the proposed work, firstly Gait is captured using moving video and binary silhouettes are generated. Secondly, feature extraction from each silhouette is done using GPPE. Lastly, recognition is performed using SVM, K-Means and MDA. Here, all the work will be performed on live input videos and gait MATLAB database.