Biometric Identification on the Basis of BPNN Classifier with Other Novel Techniques Used for Gait Analysis
Ira Gaba1, Paramjit Kaur2

1Ira Gaba, Department of Computer Science, Indo Global Group of Colleges, Abhipur (Punjab), India.
2Asst. Prof. Paramjit Kaur, Department of Computer Science, IGEF. (Punjab), India.

Manuscript received on 21 September 2013 | Revised Manuscript received on 28 September 2013 | Manuscript published on 30 September 2013 | PP: 137-142 | Volume-2 Issue-4, September 2013 | Retrieval Number: D0815092413/2013©BEIESP
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (

Abstract: Gait is the manner of the limb movement or the manner of moving a foot of an individual and recognition of an individual is the task to identify a people. Gait recognition is the biometric process by which an individual can be identify by manner they walk. The advantage of gait over other biometric traits such as face ,iris ,and fingerprint etc is that it is non-invasive and less unobtrusive biometric, which offers to identify people at the distance, without any interaction from the subject or at low resolution. In this paper, firstly the input video is converted into frame then, binary silhouette of a walking person is detected from each frame. Secondly, feature from each frame is extracted using image processing operation. Here distance between head and feet, distance between both hands, length of one hand, length of leg etc using hanavan’s model are taking as key feature. And then CBIR method is also used. At last BPNN+MDA and BPNN+LDA techniques are used for training and testing purpose. Here all experiments are done on gait database and input video. Therefore, by using the combination of BPNN with LDA and MDA, in this paper, it obtains the better accuracy results.
Keywords: BPNN, CBIR, Feature Extraction, Gait Recognition, LDA, MDA, PCA, Silhouette Extraction.

Scope of the Article: Biomedical Computing