Research On Image De-noising And Enhancement Based On Wavelet Transform

Currently, the image information is an indispensable media and the way to obtain and transmit information. However, in the practical applications, due to imperfect imaging system and the presence of transmission medium, an image is susceptible to noise interference in the process of generation and transmission, etc. And the quality of the image information will have a certain degree of reduction. Therefore, image processing, the image signal de-noising and enhancement are critical processing steps. The wavelet transform can effectively overcome the limitations of the Fourier transform in dealing with the complex image processing non-stationary signals due to the own time domain and frequency domain localization properties.And it has become an important means of image processing.In this thesis, the direction is the use of wavelet analysis method for image de-noising and enhancement. The main works and originalities are as follows:Firstly, describe the basic concepts of wavelet theory and application in image processing. Describes the impact and role of image de-noising and image enhancement technology and analyzes the characteristics and advantages and disadvantages of various algorithms in image processing.Secondly, the study about the image signal de-noising method based on wavelet theory. By analyzing wavelet coefficients characteristic of the image, in-depth studies the wavelet thresholding method selected threshold function and threshold, and proposes an improved threshold function and threshold method. Then uses the improved algorithm to remove pixels by Gaussian white noise pollution, and can maximize the retention of the image information. After the test results show that the improved algorithm on the removal of Gaussian white noise and objective indicators than the soft and hard threshold function better in the visual effects.Finally, analysis of wavelet analysis. Analyzes the histogram equalization method and wavelet feature enhancement, and image enhancement algorithm has been optimized by combining the two methods study to better highlight the image brightness and detail. And it shows that the algorithm feasibility and superiority through simulation experiments.This thesis mainly studies the theoretical basis of image de-noising and enhancement technology based on wavelet transform, and proposes a new wavelet image de-noising and enhancement method. Realizes the image de-noising and enhancement. The results verify the feasibility and effectiveness through a comparative analysis of simulation.