Adaptive Image Denoising Method Based On Wavelet Transform

In the process of collection, transmission and other processes, images are generally contaminated by noise because of various factors. The noise not only results in the decline of the image quality, but also covers up the feature in the image. Therefore, image denoising is an important step in the area of image processing, and is of great practical significance to the continuous processing of images.Wavelet transform and partial differential equation are very effective methods in the field of image denoising. Wavelet transform is widely used in the field of image denoising because of the good characteristics of multi-resolution, energy compaction and local properties in time and frequency domain. The anisotropic diffusivity of partial differential equation obtains a better denoising performance.The development of image denoising and the applications of wavelet transform in image denoising are introduced at first. Then, wavelet analysis and present image denoising methods are described and analyzed in detail. Three denoising algorithms based on wavelet are given after deeply study on wavelet theory and different denoising algorithms. They are wavelet adaptive denoising based on multiple images, multi-threshold image denoising based on undecimated wavelet transform and survey of a hybrid image denoising algorithm based on wavelet shrinkage and TV diffusion. The first algorithm makes full use of weighed computation based on multiple input images after wavelet transform, Then the threshold which is adaptive to the scale and the size of the sub-band is adopted in the denoising. The second algorithm employs the relationship between the coefficients and classifies the coefficients into three groups after the wavelet transformation, then different methods are employed to deal with the different groups of coefficients in order to improve the performance of the denoising. The third algorithm proposes a hybrid image denoising algorithm based on wavelet shrinkage and TV diffusion owing to the disadvantage of TV diffusion.Simulation results show that these three image denoising algorithms proposed in the paper are effective, obtain better visual effect, and their PSNR are also greatly enhanced.