Classification of unsharp pixels using Gaussian scale mixture (GSM) model

The aim of this study is identification of the motion blur from an given image. In large number of applications, the images are blurred by the relative motion between the camera and the object scene. The estimation of the blur kernel is highly dependent on the motion blur for the restoration of the degraded image, Furthermore to estimate these parameters of the blur kernel, we introduce the Fourier-Random Transform. We propose the solution to this problem by identifying the important parameters with which to characterize the PSF of the blur, given only the blurred image itself.