基于自适应低通滤波的超声医学图像增强算法  

A method of medical ultrasonic image enhancement based on self-adaptive low pass filter

作　　者：王绍波[1] 郭业才[2,3] 王帅 

机构地区：[1]安徽理工大学医学院,安徽淮南232001 [2]电气与信息工程学院,安徽淮南232001 [3]南京信息工程大学电子与信息工程学院,江苏南京210044

出　　处：《中国医学影像技术》2009年第3期492-495,共4页Chinese Journal of Medical Imaging Technology

基　　金：全国优秀博士论文作者专项资金(200753)

摘　　要：目的介绍一种超声医学图像增强的有效算法。方法基于自适应低通滤波器的超声图像增强算法,首先采用对数变换的方法将超声医学图像中存在的乘性噪声变为加性噪声;再通过低通滤波器将对数图像分为高频分量和低频分量,对低频分量采用自适应直方图均衡处理,对高频分量进行加权;然后对低频分量和高频分量进行融合得到增强的对数图像;最后对对数图像进行指数变换得到输出图像。结果原超声医学图像得到有效增强,边缘细节得以保留。结论该算法有效地实现了超声医学图像增强,突出了超声图像的细节,改善了视觉效果,并对噪声具有良好的抑制作用。Objective To propose an efficient method of medical ultrasonic image enhancement. Methods In this method, logarithmic transform was first carried out to the medical ultrasound image. Multiplieative noises were transformed into addi rive ones. The high and low frequency parts were then departed with the help of low pass filter. Then the low frequency component was processed with self-adaptive local area histogram equalization algorithm and the high component was weighted. Finally the two components were reunited to get the enhanced logarithmic image, the exponential transform was processed. Results The original medical ultrasonic image was efficiently enhanced and the image edge details were reserved. Conclusion The experiment results show that the particular information extrudes, the whole visual effect is improved after the process, and the method is efficient to denoise speckle noises.

关 键 词：超声医学图像 低通滤波器 自适应邻域直方图 图像增强 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术] R445.1[自动化与计算机技术—计算机科学与技术]