Image Deblurring Based on Fuzzy Kernel Estimation in HSV Color Space

Abstract

Image deblurring is intended to restore the clear images from the damaged images. The main factor of the image deblurring is to precisely estimate the fuzzy kernel in the unknown blurring process. For color images, the fuzzy kernel is estimated by converting them into the gray domain with the most existing effective image deblurring approaches. In fact, the influence of the fuzzy kernel function on each channel in the HSV color space is different. In this paper, a more accurate image deblurring algorithm is proposed. In HSV color space, the Radon transform based on the spectral edge detection is used to estimate the fuzzy kernel. For the fuzzy kernel obtained from the different channels independently, the deblurring restoration based on Richardson-Lucy guided filter is carried out in each channel. The Simulation results demonstrate that the proposed method is effective.