一种半监督的彩色图像分割方法

提出一种基于半监督EM聚类的彩色图像分割方法, 算法利用了有限的人工信息, 即在图像上点击有限的几个点以标识对应区域之间的关系, 从而得到满足给定限制的精确图像分割结果。算法首先对图像进行量化处理, 而后在量化后的色彩空间中集成先验的分割信息进行色彩聚类。实验表明算法运行速度快, 分割效果好, 具有很高的应用价值。

A Semi-Supervised Color Image Segmentation Method

A new image segmentation algorithm based on semi-supervised EM clustering is proposed, which makes use of limited human assistance, users indicate the relationship of some different regions in the image by mouse, to get the finally accurate segmentation result. The algorithm first has the image quantized and then clusters in the quantized color space with prior segmentation information. Experiment results show that the proposed algorithm is effective and has high value of utility.