The Application Of Wavelets In Medical Images

As a powerful tool for time-frequency analysis, wavelets has been widely used in many engineering fields. This thesis is mainly about the application of wavelets in medical image processing, including image denoising and segmentation. In the first chapter, the theory of medical imaging is introduced. The imaging principle of some medical equipment is also mentioned. Then the meaning and means of medical image processing are briefly explained. Noise which affects the medical image is discussed in chapter two. Meanwhile, the statistic model of many kinds of noise is also listed. The effect of the noise on some typical medical images can also be found in this chapter. The third chapter is the detailed description of the wavelets and some theory related, such as orthogonal wavelets, biothogonal wavelets and redundant wavelets. The means of implementing is also mentioned. The advantages of multi-resolution analysis in image denoising are introduced with many words. The discussion of medical image denoising by wavelets thresholding can be found in chapter four. The algorithms for deciding the threshold are also showed. Afterwards, the mixed algorithm of wavelets and watersheds, which is used for the medical image segmentation, is introduced in chapter five. The algorithm based on marker-controlled watersheds segmentation is merged with the wavelets. The experiment shows that this algorithm is effective for the medical image segmentation. The chapter six shows some experiment result. The algorithms mentioned in this thesis are effective in the medical image processing, either in denoising and segmentation of CT images. The chapter seven is the conclusion of this thesis and some work to do in the future is also discussed.