The Restoration From Focus Fuzzy Image With Noise

With the development of computer technology and product marketing, the digitization of images and popularization of display equipment are widely applied in military detecting, communication and transportation, artificial intelligence, astro observation, remotely-sensed image, satellite observation, social production, medical facility and so on.Image blurring and noise effect are the two important factors which can cause the image degradation in the acquisition, transmission, compression and storage of image that makes the denoising and restoration of image come to an important area in digital image processing.This paper researched the restoration of image with fuzzy noise based on the characteristics of images, and extended the algorithm which can be applied in the restoration of image with different depth of field. What we did are as follows:Firstly, the basic theory, knowledge and forming of blurred image, the different of blurred reasons and Fast Fourier transform are introduced. The optics transfer function is elicited with the theory of optical imaging next.Secondly, different noise and different ways to denoising are introduced. What we pay more attention on the knowledge of different filtering methods such as average filtering and median filter. The advantages of algorithm in experiments are given by the analysis of the correctional method of average filtering and median filter.Thirdly, the classical restoration algorithm of inverse filtering, wiener filtering, Lucy-Richardson, least square method and maximum entropy in focal fuzzy image are introduced, and their advantages and disadvantages are also pointed out. Then, we introduce blind restoration algorithm of focus fuzzy image by computing the point spread function, and indicate what algorithm is used in this paper.Finally, a improved algorithm is proposed and implemented using matlab, which is based on the mastery of the least squares algorithm and adaptive reconstruction algorithm. At the same time, the advantage of the new algorithm is proved using the simulation experiment, and the universality of the algorithm is validated by covering the focus fuzzy image which is taken with camera.When we take photos, image fuzzy degree varies with the depth of field. So, the image block recovery method, which is based on image segmentation is not only introduced, but also implemented using matlab in this paper. In order to verify the feasibility of segmentation, the segmentation image of focus fuzzy image, clear image and local blurred image is presented. What’s more, the cockiness piece of restored image of local fuzzy is also given to present the advantages of the sectioned restoration algorithm.