Image Restoration For Motion Blur Based On The Normalized Sparse Prior And Guided Filter

With the continuous development of the times, people’s requirements become increasingly high on the image quality. Image restoration technology as a method to improve the quality of the image which has been paid attention to. Image deblurring has been widely used in various fields in life, such as: aerospace remote sensing, bio-medical, industrial. Therefore, the study of the image restoration has a very important practical significance.However, in the process of acquiring an image, blurred image will appear when there is a relative position between the camera and the scene. The blurring process is represented by the point spread function, but in the point spread function is unknown, image de-blurring into a serious ill-posed problem, which greatly increases the difficulty of processing.This paper mainly deals with motion blur processing to single blurred image as an input image, the point spread function in unknown conditions, the blurred image recovery process. First, detailing the research status at home and abroad in the field of image restoration. Second, the relevant theoretical knowledge in the field of image restoration, including Fourier transform, image degradation model, some of the image quality evaluation methods. Third, we introduce some of the common image restoration algorithms, such as inverse filter, wiener filter, Richardson-Lucy.Finally, a based on image normalization sparse prior method is proposed to estimate point spread function, and a based on the guidance of the filter method is proposed to restore image.Experimental results show that the algorithm processing motion blur images can get a good restoration result. On the one hand, the point spread function estimation is or not accurate in the point spread function is unknown, direct impact on the results of the recovery image, in this paper, the point spread function of the proposed estimation algorithm can accurately estimate the point spread function.On the other hand, restoration algorithm based on the guided filter is proposed in this paper, which is divided into two parts de-noising and de-blurring, which can achieve good noise suppression effect when a blurred image contained large noise.