基于形状参数的重叠细胞图像判别研究

Overlapping Cell Images Discrimination Analysis Based on Shape Parameter

利用计算机技术分割重叠细胞图像是目前图像处理中重要一部分，对诊断一些疑难杂症具有重要意义。而目前存在的分割重叠细胞图像大部分都直接对图像进行分割，容易造成单细胞被分割现象，同时降低效率。该文在对重叠细胞图像进行分割之前，首先利用形状参数时其进行判别，从而得到一幅只剩重叠细胞的图像，有效避免误分割现象。

It is an important part for using computer technology to separation overlapping cell images. And it is also important sig- nificance to diagnose difficult miscellaneous diseases. While the existing segmentation of overlapping cell images most directly on the image segmentation. So that single cells are fractionated phenomenon easily, and reducing the efficiency. It is firstly using a shape parameter to discriminate overlapping cell images before segmentation. Thus getting a image only overlapping. It effectively avoid false segmentation.