一种基于小波变换的医学图像增强新算法

针对目前的增强算法对噪声比较敏感的特点,本文提出一种基于多尺度小波模值的对比度增强新算法.通过设定不同的模值拉伸因子,改变不同尺度下的小波系数的模值,来增加图像反差,增强边缘等特征细节信号.同时利用信号与噪声的Lipschitz指数在局部奇异处呈现不同的表现形式的特性,滤除噪声信号,达到去噪和特征增强的双重目的.实验结果表明,该算法对噪声有一定的抑制作用,可以在提高图像对比度的同时滤除噪声信号,有效地解决了传统方法中存在的强去噪能力和高对比度增强之阃的矛盾.

Contrast enhancement algorithm for medical images based on wavelet transform

Because of the sensitivity of present enhancement approach to noise, an improved enhancement algorithm based on multi-scale wavelet modulus is presented. By proper selecting different stretching coefficient, and changing wavelet coefficients modulus in different scales, contrast is enhanced and detail information such as edge is intensified. Furtherover, noise in images is filtered by using the different characteristics of Lipschitz index between the signal and noise in local singularity points. Two aims are realized:the noise is reduced and the edge of the image is enhanced. The results show that this method can inhibit the noise at certain extent. In virtue of this new algorithm, detail features is enhanced and the noise is attenuated at the same time. The algorithm can resolve the inconsistency between high contrast enhancement and high denoising ability existed in traditional algorithm.