细胞形态参数的准确获取对于临床诊断具有重要的意义。本文基于图像处理技术,采用MATLAB图形用户界面(GUI)开发了一个自动获取AFM细胞形态参数的软件平台,包括图像增强、图像分割、图形修饰、形态参数计算等功能。针对常见的细胞粘连现象,提出了一种改进的分水岭算法,即先将二值化后的图像进行距离变换,然后采用形态学开重构运算对距离变换图进行重构,最终用标准分水岭算法分割重构图。实验结果表明,该方法能够有效防止过分割现象。此算法不仅对细胞图像有效,对其它近似圆形的颗粒图像亦适用。
Acquiring cell's morphological parameters has important significance for making clinical diagnosis. In this paper a new software for automatic acquiring morphological parameters of AFM imaged cells was developed based on image processing technology and MATLAB graphical user interface (GUI). The software included image enhancement, image segmentation, image modification and morphological parameters calculation. In order to solve the problem of segmenting overlapping cells, an improved watershed algorithm was presented. Firstly, distance transform was applied to the binary image. Secondly, opening-by-reconstruction operators were employed to reconstruct the distance transform image. Finally standard watershed transformation was used to segment the result image. The experimental results show that the algorithm can effectively prevent the phenomenon of over-segmentation. The algorithm not only can be used for cell images, but also other kind of disk-shaped granules images.