A review on various state of art technique to recognize occluded face images

One of the well-known problems of the face recognition is occlusion. Occlusion in an image refers to hindrance in the view of an object. The article aims to give a detailed survey of the face recognition under occlusion. Human face recognition under occlusion is broadly classified into 8 categories Karhunen-Loeve Expansion Method, Model Based Method, Correlation Based Method, Template Based Method, Feature Based Method, Geometric Based Method, Singular Value Decomposition Based Method and Neural Network Based Algorithm. For these categories standard databases used are AR, Ex. Yale B, FRGC v2, CMU Multi-PIE, Bosphorus, CAS-PEAL, LFPW, FERET. Structured Sparse Error Coding method using Extended Yale B Database gives the recognition rate of 99% which is the better method compared to other methods. A comparative analysis helped us to know the merits and demerits of the referred articles.