Face Verification Across Age Progression with Wavelet Transforms: A Review

Abstract In human faces, a lot of change in the appearance is seen as their age is increased. In order to recognize faces which are under varying illumination, we have proposed a wavelet-based normalization method so as to normalize the images. This method will normalize the face for extracting and matching the images with the stored images in the database. In this research paper for normalizing the image we are using discrete wavelet transform (DWT) and for matching the images with the stored image we are using minimum distance feature vector algorithm and template matching method. It will calculate the matching percentage of the original image with the new image. Again, the proposed research work for this paper is to compare the matching percentage with the wavelet transform and the minimum distance feature vector algorithm. For taking the dimension of that image again we are using the Gradient Orientation Pyramid and Support Vector Machines (SVM’s).