AVERAGE-HALF-FACE FOR FACE RECOGNITION USING PCA AND WAVELET TRANSFORM

Abstract

With the tremendous development in technology, security s ystem has also reached its peak in development. Biometrics has played an important role in security system in the past decades. As the most natural and friendly identification method, automatic face recognition has become the important part of the next generation computing technology. 3D face recognition methods are able to overcome the problems resulting from illumination, expression or pose variations in 2D face recognition. Many face recognition algorithms have been developed and each has its strengths. Each shares a common element which is that they all input a full face image into the algorithm. They need more memory space and take large computation time. Utilize the symmetry of the face for face recognition, select average half-face for research. The consequence of this discovery may result in substantial savings in storage and computation time. We have good response on the average half face image using wavelet transformation; reduce the computation time for each image is one third of the full image. And also we determined the memory storage is reduced the half of the full image. We save memory space for each image is 60KB. Applying the average-half-face to additional algorithms and databases (2D&3D) and analyzing the effect of illumination, facial expressions and occlusions. A face recognition method that is able to recognize faces at various angles is proposed.