基于分水岭变换的粘连颗粒图像分割方法

提出了一种基于分水岭变换的粘连颗粒图像分割方法. 首先对图像进行预处理，进行二值化；然后通过距离变换和灰度形态重构得到每个目标的种子区域（目标标记）；再根据目标标记使用强制最小技术修正距离变换图；最后，对修正后的距离变换图进行分水岭变换，得到分割结果. 在Matlab环境下进行实验，结果表明该算法效果良好，能有效的抑制过分割.

Method of the Segmentation of Touching Image Based on Watershed Transform

In order to realize the touching image segmentation, we used the watershed transform. The first step is the image pre-processing. The main purpose of this step is to transform the original image into a black-white image. The second step is to get the foreground marker using Euclidean distance transform and morphological reconstruction. The third step is to modify the distance transform image using impose minima technology. The last step is the watershed transform. In order to test the new watershed algorithm, we did some experiments using Matlab, and the results show that the watershed algorithm does well and can prevent the over-segmentation.