血管内超声图像的噪声抑制与增强算法研究  

Algorithm for noise suppression and image enhancement based on intravascular ultrasound images

作　　者：王志东[1] 汪友生[1] 李龙[1] 董路[1] 李冠宇[1] 

机构地区：[1]北京工业大学电控学院,北京100124

出　　处：《电子测量技术》2013年第3期44-47,共4页Electronic Measurement Technology

摘　　要：针对含强背景噪声的血管内超声图像,噪声滤波和边缘增强是对血管内超声图像进一步处理的基础工作。提出一种自适应中值滤波结合改进的高通滤波的方法,有效的滤除了噪声并保持了边缘信息。Retinex增强算法可获得图像中各像素的相对明暗关系,并通过这种相对关系进行线性映射,从而增强图像且对噪声影响较小。实验结果表明,综合运用该滤波和增强方法,在滤除噪声的同时对边缘的增强有着比较满意的效果,为以后血管中外膜的提取提供了基础。According to intravascular ultrasound （IVUS）images with strong background noise, noise filtering and edge enhancement are the basic work of further processing on IVUS images. An adaptive median filter method combined with an improved high-pass filter is proposed, which effectively filters out the noise and keeps edge information. The Retinex enhancement algorithm can obtain the relative darkness relation between each pixel in image and conduct linear mapping through this relation,which can enhance the iraages and has a little influence on noise. The experimental results show that the integrated using of this filtering and enhancement method has a satisfactory effect on filtering noise and enhancing the edge,which provides a basis for media-adventitia border detection in the future.

关 键 词：血管内超声 自适应中值滤波 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]