Algorithm for Face Matching Using Normalized Cross-Correlation
C. Saravanan1, M. Surender2
1Dr. C. Saravanan, Head, Computer Centre Dept, National Institute of Technology, Durgapur, West Bengal, India.
2M. Surender, Information Technology Dept, National Institute of Technology, Durgapur, West Bengal, India.
Manuscript received on March 02, 2013. | Revised Manuscript received on April 13, 2013. | Manuscript published on April 30, 2013. | PP: 930-934 | Volume-2, Issue-4, April 2013. | Retrieval Number: D1388042413/2013©BEIESP
Open Access | Ethics and Policies | Cite
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Face matching is the most important and crucial procedure in face recognition. It is difficult to achieve robust face matching under a wide variety of different image capturing conditions, such as lighting changes, head-pose or view-angle variations, expression variations, etc. Robust face matching is essential to the development of an illumination insensitive face recognition system. This paper proposes a face matching algorithm that allows a template called extracted face of person which is the Region of Interest from one image and start search for matching with the different image of same person taken at different times, from different viewpoints, or by different sensors using Normalized Cross-Correlation (NCC). The algorithm is implemented in MATLAB. The experimental results show that developed algorithm is robust for similarity measure.
Keywords: Face Matching, Normalized Cross-Correlation (NCC), Region of Interest (ROI).