基于流域算法的粘连颗粒图像分割

在分割颗粒图像中,经常会有颗粒重叠粘连,这给图像的后续分析处理带来了很大的麻烦。针对这一问题,提出了一种有效的基于流域分割算法的颗粒图像分析方法。该算法提出一种新的种子点选取方法,使得种子区域最大化,然后以种子区域为核心通过流域膨胀变换重建原图,当图像中所有像素点都重建完时,粘连颗粒图像的分割就完成了。文中主要叙述了算法的设计思想及实现,并给出了实验结果。实验表明该算法能有效改善因多种子点而引起的过分割现象,对形状不规则的颗粒也同样适用,能取得比较满意的分割效果。

Separate Algorithm for Touched Particle Images Based on Watershed Algorithm

In the particle images segmentation,the objects are always touched or overlapped each other,which makes it hard for image consequent processing.In order to solve the problem,an effective separate algorithm based on watershed algorithm is proposed.First,adopt a new method to get the maximum seed regions.Second,reconstruction the particle images by watershed transform,dilation from the seed regions.We can finally to separate the touched particle image when all the image pixels have been assimilated to a separ...