改进分水岭算法与K-means方法结合的图像分割

Combining Improved Watershed Algorithm with K-means Method in Image Segmentation

针对分水岭分割算法存在的过分割及对噪声敏感问题,提出一种基于K-means聚类算法与改进分水岭算法结合的图像分割算法,首先,利用K-means聚类算法进行初始聚类分割,提取感兴趣的目标;然后,提出基于4-邻域相似度的改进分水岭算法,对K-means初始聚类图像应用改进分水岭算法分割目标区域。从100幅人骨医学图像提取人骨区域,实验结果表明所提出算法可解决分水岭算法的过分割问题,且有效分割了图像目标。

In view of the over segmentation and sensitivity to noise problems of watershed segmentation algorithm,an image segmentation algorithm was proposed based on improved watershed algorithm and K-means clustering algorithm.Firstly K-means clustering algorithm was used for the initial clustering segmentation.Then,an improved watershed algorithm based on the 4-pixel neighborhood similaritywas proposed,and the improved watershed algorithm was applied to the Kmeans initial clustering image to segment the target area.The bone regions were extracted from 100 medical images of human bones,and the experimental results show that the proposed algorithm can effectively solve the over-segmentation problem of watershed algorithm,and effectively segment the image target.