The Study Of Medical Image Enhancement Based On Wavelet

In recent years, wavelet image processing has been used in many fields. It canselect appropriate wavelet methods for image processing according to differentapplication needs. In medicine, it often requires high-quality images to help doctorsdiagnose diseases. However, due to the restrictions of real medical conditions, theimages obtained in the medical process usually cannot meet the actual demand.Therefore, using wavelet analysis and image processing techniques to enhance theprocess of image processing, can improve the accuracy of medical diagnosissignificantly.For image processing, most of the wavelet algorithms are decomposing the imagesinto multi-resolution, and they mainly deal with the low frequency sub-images thatobtained from the decomposition, but recent studies show that the processing of highfrequency sub-images is also effective. This paper proposed an algorithm for enhancingand denoising the high frequency sub-images, and for edge enhancement of the lowfrequency sub-images, and then designed the relevant gain weight function combiningwith the Mammography X-ray.The advantage of wavelet image processing is the disposition of points, but theextracting effect of edge information (e.g. tumor) in the images is not very good, whilethe curvelet transform has very obvious advantage for this. This paper improved theabove algorithm, and used the second generation curvelet transform to improve thealgorithm for edge extraction process, which has more obvious advantage than thetraditional curvelet transform in practical application.At last, this paper used Matlab to make a simulation experiment for the medicalimages. The experimental results of visual effect and image indicator data show that theimproved algorithm has many advantages compared with the traditional enhancementalgorithm, such as the texture of the images processed by the improved algorithm ismore clear, calcification pots are independent, the edge area is clear, the amount of thecalculation is small, and the most important is that it will be more helpful for diagnosisof diseases.