Ear recognition under partial occlusion based on sparse representation

Current research on ear recognition in 2D achieves good performance in constrained environments. However the recognition performance degrades severely under occlusion, noise or pose, illumination variations. This paper proposes a 2D ear recognition approach based on sparse representation to deal with ear recognition under partial occlusion. Firstly, the ear part is automatically detected and extracted from the source image. Then, we use different methods (down sample, PCA, LDA and random projection) for feature extraction. Thirdly, sparse representation classifier is applied for ear recognition under occlusion. Experimental results on the USTB ear dataset verify the efficacy of the proposed method.