Application Research of Gabor Filter and LPP Algorithms in Face Recognition

摘要

This paper propose an optimized algorithm for face recognition based on Gabor filter and Locality Preserving Projections (LPP) Algorithm, improve the high dimensions problem of face recognition use Gabor filter. The Algorithm is as follows: First, Use Gabor filters to decompose face images to get a high-dimensional vector; Second, Reduce the dimension of high-dimensional vector by LPP; Finally, The obtained feature vector is input to SVM to efficiently identify the face. The proposed algorithm is test by ORL face database. Experimental results show that the proposed algorithm can efficiently identify the face, and the overall performance is better than the algorithm, which is based on only Gabor filter.