Estimating Motion Blur Parameters with Gradient Descent Method

This paper proposed a method for motion deblurring by estimating parameters length and angle in Point Spread Function (PSF) with the gradient descent method based on an absolute gradient magnitude. These parameters are used to restore the blurred image to be sharp by using wiener filter. The experimental results show that the proposed method is effective for estimating parameters and improves a quality of the blurred image. Peak Signal to Noise Ratio (PSNR) and Structural Similarity index (SSIM) are applied to test the efficiency of the method.