Image Denoising Based On Dual Tree Complex Wavelet

Along with scientific and technological advances, network and computer have become more and more popular in our daily life. Various forms of multimedia files greatly enrich people’s lives. Image with its information as well as the transmission speed has become an important source of information. However, during the process of acquisition and transmission of image signal, noise signal will affect the image with a variety of disturbances, which will greatly reduce the quality of image signal. And it will also have a tremendous negative impact on the processing, transmission and storage of image signal. Therefore, image denoising has great value of application.Traditional image denoising was mainly conducted within the spatial domain or frequency domain. A variety of filters are used for image denoising such as mean value filter, median filter, Wiener filtering and denoising method based on Fourier Transforms. But these methods, for image signal processing or completely in the space at all or in the frequency domain, for image signal, the frequency characteristics tend to change with the changes in spatial location. Wavelet theory solves the problem. Currently, image denoising based on wavelet analysis technology has played an important role in the field of image denoising.The methods of wavelet-based image denoising are studied in this thesis. An efficient method of image denoising is presented by introducing dual-tree complex wavelet. This method is compared with the traditional methods by simulation experiments to realize the image denoising.