基于改进分水岭算法的组织细胞图像分析  

Watershed-based Segmentation of Histiocytic Images

作　　者：赵宇明[1] 崔磊[2] 柴岗[2] 吴越[1] 朱锴[1] 

机构地区：[1]上海交通大学图像处理与模式识别研究所,上海200030 [2]上海组织工程研究与开发中心,上海200011

出　　处：《生物医学工程学杂志》2005年第6期1151-1156,共6页Journal of Biomedical Engineering

摘　　要：组织细胞图像分割算法的研究是组织学图像检测技术的关键和难点所在。本文提出的细胞分割算法首先运用基于数学形态学的预处理把图像灰度间距拉开,然后模拟梯田灌溉的方式,通过先过分割再合并的改进分水岭算法来分割组织细胞图像,以得到最终结果。针对骨髓和软骨细胞进行实验,实验结果表明该算法的有效性。The task of segmenting histiocyte is a crucial step in the analysis of histiocytic images which is an important application of computer vision to histopathology. The algorithm presented in this article was composed of two steps： （1） the morph-based preprocessing; （2） the ameliorated watershed method. In the first step, the difference between histiocytes was magnified in order to increase the visibility from the view of the computer vision, and then the ameliorated terrace-flooding-simulated watershed method was used to achieve the segmentation of histiocytic images in the second step. To test the performance of the algorithm, different samples of visual quality were tested and the result figures proved successful.

关 键 词：数学形态学 分水岭 组织细胞 分割 

分 类 号：R318[医药卫生—生物医学工程]