The blur function of a degraded image is often unknown a priority. The blur function must first be estimated from the degraded image data before restoring the image. We propose an algorithm to address the blur estimation problem. The present algorithm based on the estimation of restoration filter parameters by using edge information of the degraded image is presented to solve the restoration problem of the degraded image. The information that relates the variance of the Gaussian blur kernel on degraded image is considered. Simulation results of image restoration illustrate the performance of the proposed estimation method.