Research On The Astronomical Image Restoration Algorithms For Mixed Degradation Model

Image is ofen disturbed by many factors in the process of obtaining, and it is hard to avoid dropping mass or degradation. Image restoration is an important problem in image processing technology, and it has very important meaning for improving image quality. The key of solving the problem of image restoration is using the prior knowledge of image degradation process to establish the mathematical model of image degradation process, then solve this inverse problem to get a restoring model and an estimation of original image in a reasonable manner. Image restoration is the basis of image processing, pattern recognition, machine vision, etc., and it has been widely used in the fields of astronomical observation, space exploration, medical imaging, military science,criminal investigation, traffic monitoring and other fields.In this paper, we study the application of image restoration in astronomical observation and discuss the astronomical image restoration methods.The main work of this paper includes:(1)Introducing the concept of image restoration, causes of image degradation, noise and denoising method, and fuzzy classification and image degradation model. (2) Introducing four classic image restoration methods and blind image restoration method in detail,and using inverse filtering, wiener filtering, least squares filtering to restore image respectively based on the motion blur degradation model and the turbulent fuzzy regression model, and it is concluded that L-R algorithm will have the best restoration result under the condition of knowing point spread function (3) Respectively using point spread function of motion blur degradation model and turbulence blur degradation model as the initial point spread function of the blind image restoration method to restore image based on mixed degradation model (motion blur degradation model and turbulent blur degradation model), and it is concluded that using point spread function of turbulence blur degradation model as the initial point spread function of the blind image restoration method will have a better result, therefore, the point spread function of turbulence blur degradation model is chosen as the initial point spread function (PSF) for astronomical image blind restoration. (4) Using approximate point spread function of blind image restoration method for image restoration to guid L-R algorithm restoring image again,and finding that the results is better.This paper uses the Matlab simulation experiment, the experimental results show that the fuzzy degradation of the astronomical image degradation model conforms to the assumption of mixed degradation model (motion blur degradation model and turbulent fuzzy regression model), and result of using blind image restoration method and L-R restoration algorithm to restore image is increased.