Architecture for precisive face recognition system deploying quantization with multiresolution curvelet and training with support vector machine

摘要

The proposed research approach addresses various issues in face recognition as well as in computer vision which were signified and researched by multiresolution concepts like wavelet transforms. But survey also shows that only wavelets are not the factors for ideal description of an image as they have very rough directional representations and are absolutely not anisotropic. With the recent developments in Curvelet Transform, which has the potential to overcome these flaws of wavelets, this proposed idea highlights an advance technique of face recognition which is based on multiresolution curvelet. The application will quantize 8 bit image to 4 bit and 2 bit representation in initial phase. In the next phase, the curvelet transform will be applied to all 3 different resolved versions of the image. In the final phase, all the 15 sets of co-efficients were used to train different support vector machines. Finally, the accuracy of the application will be evaluated by identification from different sets of facial image for the proposed robust face recognition system.