Face recognition using Eccentricity-Range based Background Removal and Multi-Scaled Fusion as pre-processing techniques

Face recognition under varying background, pose and illumination conditions is challenging, and extracting the corresponding invariant features is an effective approach to solve this problem. In this paper, we propose two novel preprocessing techniques, viz., Eccentricity-Range based Background Removal and Multi-Scaled Image Fusion, to improve the performance of a face recognition system. Eccentricity-Range based Background Removal is used to ascertain the shape of the face and distance from the camera to eliminate complex backgrounds. Multi-Scaled Image Fusion is used to neutralize the effect of pose. The resulting pre-processed image contains the salient edge details of the face and prepares the ground for DWT based feature extraction. Experimental results show the promising performance of the proposed techniques for face recognition on two standard face databases, namely, Color FERET and CMU-PIE.