Face Recognition System using EBGM and ANN
Akhil Mahajan1, Parminder Kaur2

1Mr. Akhil Mahajan, Department of ECE., Chandigarh Engineering College, Mohali (Panjab), India.
2Ms. Parminder Kaur, Associate Prof., Department of ECE., Chandigarh Engineering College, Mohali (Panjab), India.

Manuscript received on 21 September 2013 | Revised Manuscript received on 28 September 2013 | Manuscript published on 30 September 2013 | PP: 14-18 | Volume-2 Issue-4, September 2013 | Retrieval Number: D0753092413/2013©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 (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Abstract: There are many challenges associated with face recognition systems which make them a complex and difficult process. These factors include—pose variations, facial expressions, occlusion, age etc. These factors affect the face recognition systems and deteriorate their accuracy. The face recognition problem can be solved by using some statistical techniques such as PCA, ICA and LDA. Some feature based techniques—Elastic Bunch Graph Matching (EBGM), Artificial Neural Network (ANN), etc. have also been used and implemented to solve the face recognition problem. In this paper an insight is provided into various techniques available for face recognition, and a method is proposed that provides an efficient and feasible solution for real-time face recognition system. The proposed method uses EBGM technique, which in turn uses facial features for the identification of the test images that may be captured from a live video. Experimental results show that by involving ANN, better matching results with EBGM were obtained. Moreover, for face recognition in live videos and under low illumination conditions, the proposed system works more efficiently and gives better matching results when compared with the other techniques.
Keywords: Elastic Bunch Graph Matching (EBGM), fiducial points, Independent Component Analysis (ICA), jets, Linear Discriminant Analysis (LDA and Principal Component Analysis (PCA).

Scope of the Article: Pattern Recognition