摘要
文章提出一种无信息丢失的图像处理方法,用以校正受到生物组织随机蠕动影响的光学相干层析(OCT)图像。文中首先利用相干增强各向异性扩散算法对原始OCT图像预处理增强层状结构,再利用顺序查找峰值的方法找出某层状结构及其周围层状结构的位置,进而利用异常点去除的方法对该层状结构进行校正,并以校正后得到的该层状结构位置作为基准校正图像。对眼底OCT图像去抖动的实验说明该方法能很好的去除OCT图像中的抖动,从而很好地辅助诊断。
An image processing method without information loss is presented in this paper, and the method is applied to optimize OCT images affected by the random moving of tissue. First, layer structures are enhanced by using coherence enhancing diffusion algorithm. Then sequential peak finding algorithm is used to find the location of the specific layer structure and its surrounding layer structures. After that, abnormity removal method is used to adjust the location of the specific layer structure, and the result is used as benchmark to optimize OCT images. With OCT images of retina as examples, it is shown the above method can effectively remove dithering in these images and be clinical helpful.
出处
《激光与红外》
CAS
CSCD
北大核心
2007年第3期288-291,共4页
Laser & Infrared
基金
国家自然科学基金(60578003)
广东省攻关项目(2005B16001149)
