A novel method of image enhancement based on the fractional Fourier transform

A novel image enhancement method is presented in this paper. We apply the idea of Laplacian Pyramid and generalize it into the fractional Fourier domain to extract some image information. Firstly, we deal with the image in the fractional Fourier domain with a low-pass filter, and then the first level result can be obtained by subtracting the filtered image from the original image. The second level image thus can be received when applying the similar process to the low-pass filtered image of level 1. Further difference images can be achieved by iterating this process. Since the fractional Fourier transform has a rotation angle a, characters of image thus change in different transform domain. Image information may appear differently in every level of the difference images under some rotation angle, which makes it possible to extract image features more effectively in the fractional Fourier domain than that in the traditional Fourier domain. The proposed method provides an alternative way to enhance image with proper filter designing in the fractional Fourier transform. Simulations are provided to show the efficiency of this approach.