The Research On Algorithms Of Digital Image Restoration

There are a lot of factors such as the phase difference of the optical system, the atmosphere turbulence, moving, diffusion of the focus and the system noise that degrade the digital images during their obtaining. The destination of image restoration is to recover image that has been degraded and make sure that the processed image as near as possible to the original image. In general, in order to recovering the degraded image we resort to filtering of image. In filtering of image, both denoising and maintaining of the detail about the image is required. But in most cases denoising of image conflicts with maintaining of detail in restoration processing. For the reason that mentioned above restoration of image is very challenge as a issue of digital image processing full of importance. Many problems attached to this specific issue call for better solutions that have not been provided at present.In this paper, several algorithms dealing with restoration of digital image are proposed and discussed based on some basic classic approach. By some improvement and combination operations these new algorithms can do better than some basic classic approach in work.In introduction to this paper, an overview of some advances in restoration of digital image is recounted at first, then the construction and the organization of this paper are listed.Chapter two touches on some basic concepts in relation to digital image processing especially restoration of digital image. These concepts including the area of digital image processing, A/D and quantization of analogue image, mathematical model of noise and some parameters about estimation of objective effect of restoration result of digital image as well.Chapter three mainly deals with image restoration in the spatial domain. At first, a model of image degradation and restoration is presented. In the followed part, some basic classic noise-reduction spatial filters are discussed briefly. In the last part, four novel filters are discussed in detail. The first is an adaptive filter used in situation of blended noise. The second is an all-direction-element-based weighted average filter. The third is a quadr-state average filter. The fourth is a correlation-base stack filter. Experiments show that the first three filters are superior to the basic classic filters in image restoration. The stack filter operates quickly and it can be used under real-time condition.Chapter four is concerned mainly with digital image restoration method in transform domain. This chapter is divided into two self-contained sections oriented to filters based on Fourier transform and filters based on wavelet transform respectively. Inspired by the common minimum mean square error (Wiener) filtering and the constrained least squares filtering, a recursive blind recovery filter is derived in Fourier transform domain. Two other algorithms take advantage of the multiresolution of wavelet transform are studied in the laterpart of this chapter.All the algorithms proposed in this paper take the form of improvement by contrast with basic algorithms. To make sure that the processed image as near as possible to the original image, more research exploring faster, higher efficiency and more intelligent comprehensive algorithms of digital image restoration will take out.