Gabor texture representation method for face recognition using the angle and distance between feature points

In this paper we propose a new algorithm based on Gabor wavelets for human face recognition from a single exemplar image per person. In our approach, Gabor filters for face recognition without forming graph has been used besides information gained from angle and distance of feature points. Reasonable results are achieved by calculating difference between two images based on Gabor information, angle and distance information of feature points and combination of them in proposed method. With the Yale, AR and ORL databases, experiments exhibited that our proposed method obtained recognition rates of 88.2%, 96.5% and 82.0%, respectively.