中文摘要：

基于多尺度Retinex （Multi-scale Retinex,MSR）模型的图像增强方法是一种基于人眼视觉原理的图像增强方法,能同时有效实现图像的动态范围压缩和颜色保真.本文融合Retinex理论,采用3个尺度的Retinex算法对肝脏超声图像进行增强处理,提取经MSR算法增强处理后的正常肝脏和脂肪肝图像的灰度均值、对比度和信息熵参数,并与直方图均衡化算法、同态滤波算法进行对比.实验结果表明：肝脏超声图像经MSR算法增强处理后,提高了图像特征参数（对比度、信息熵）的区分度,增强了图像暗区的对比度和清晰度,改善了图像视觉质量,能有效辅助临床诊断.

英文摘要：

Image enhancement algorithm based on multi-scale retinex （MSR） combines the dynamic range compression with the tonal rendition at the same time.Three-scale Retinex model is applied to enhance ultrasonic liver images.Then the lightness,contrast,and entropy of the enhanced images are extracted and extraction results are compared with those of the usual image enhancement methods,such as histogram equalization algorithm and homomorphic filtering algorithm.Experimental results show that the image enhancement algorithm based on MSR model can improve the distinguish degree of ultrasonic liver image features,such as contrast and entropy,as well as increase them,which is helpful to the assisted diagnosis of liver diseases.