Image Denoising Algorithm Based On Wavelet Analysisi Theory

Since image is one of the main sources to obtain the information, the quality of the image become particularly important. However, being affected by the external factors, such as equipments and environment, the image is often accompanied by noise in the process of image acquisiton and transmission, which have a great impact on extracting useful information from the image.Therefore, it is extremely important to eliminate the noise before analyzing and utilizing.Becacuse of its good performance on time-frequency analysis, Wavelet analysis theory has been widely applied in the field of image denoising. In the paper,we present the the basic theory of the wavelet transform, study several kinds of traditional wavelet threshold denoising methods, simulate and analyze the Bayeshrink threshold denoising method using different wavelet functions and threshold functions. Based on previous research, we propose a hybrid Fourier-wavelet image denoising method based on context model.By applying this method, the noise of the smooth part of the original image was sparingly reduced in the Fourier domain at first.Then, the remaining noise was removed in the wavelet domain:A series of wavelet coefficients were obtained by using the stationary wavelet transform to decompose the image signal. Considering the correlation of the coefficients, the variance of those wavelet coefficients was calculated by the use of the context model. Then, the adaptive threshold was acquired by substituting the variance into the threshold expression estimated form the GGD model. After processing the wavelet coefficients with soft threshold function, the image denoising can be obtained by the inverse wavelet transform of the wavelet coefficients. Simulation results show that this method not only can effectively filter image noise, but also be able to keep the detail signals of the edges of the image and inhibit the Gibbs phenomenon caused by denoising. Finally, the image forming principle of the OCT system is introduced. After that, characteristics and sources of the noise of the OCT image are analyzed. Then, an OCT image was denoised by the wavelet transform method at last. The simulation results show that this method can improve the quality of the OCT images and can provide strong support to the clinical diagnosis.