基于边缘检测与分裂合并的图像分割算法

针对传统分裂合并算法容易产生方块效应与过分割的缺点,提出了一种结合边缘检测和分裂合并的图像分割算法.该算法直接利用图像的边缘信息进行分裂,不断将图像分裂为一些不规则形状的一致性区域,然后根据一定规则将相似的区域合并.实验表明,该算法能大幅减少分裂次数,并有效克服方块效应和过分割等缺点,图像分割效果较好.

An image segmentation algorithm based on edge detection and split-merge

An adaptive image segmentation algorithm based on edge detection and split-merge is presented in this paper in order to solve the problem that tradition split-merge image segmentation algorithm can easily lead to over-blocking effects and over-segmentation.The algorithm directly uses the edge information during the split process by continuously splitting image into the consistency of irregular shapes,then combines the regional shape according to certain rules.Experimental results showed that the proposed algorithm can significantly reduce the number of division and effectively overcome the box effects and over-segmentation,then obtains good effect in image segmentation.