Computer Engineering and Applications ›› 2017, Vol. 53 ›› Issue (13): 16-20.DOI: 10.3778/j.issn.1002-8331.1703-0011

Previous Articles     Next Articles

Image segmentation method based on improved similarity measure of spectral clustering

ZOU Xuhua1, YE Xiaodong2, TAN Zhiying2, LU Kai2   

  1. 1.Department of Automation, University of Science and Technology of China, Hefei 230027, China
    2.Institute of Advanced Manufacturing Technology, Hefei Institutes of Physical Science, Chinese Academy of Sciences, Changzhou, Jangsu 213164, China
  • Online:2017-07-01 Published:2017-07-12


邹旭华1,叶晓东2,谭治英2,陆  凯2   

  1. 1.中国科学技术大学 信息学院自动化系,合肥 230027
    2.中国科学院 合肥物质科学研究院 先进制造技术研究所,江苏 常州 213164

Abstract: Considering the low accuracy of image segmentation method of traditional spectral clustering, an improved similarity measure of spectral clustering is proposed. Firstly, an image is made up of some superpixels by the pre-process of superpixels segmentation algorithm, and a graph based on superpixels is constructed. Secondly, similarity matrix is obtained by the similarity calculation of superpixels, which fully considers the features of superpixels including covariance descriptor, color information, texture information and edge information. Finally, NJW algorithm is used to segment the graph based of superpixels. Compared with current unsupervised segmentation algorithm, a lot of experiment results show that the proposed approach has higher segmentation accuracy. Besides, the object marked by user can be segmented precisely using proposed approach.

Key words: spectral clustering, image segmentation, similarity measure, superpixels, covariance, NJW algorithm

摘要: 针对传统谱聚类图像分割方法存在分割准确度不够高的缺点,提出一种基于改进的相似度度量的谱聚类图像分割方法。该方法首先使用超像素分割算法将图像预分割为一定数目的超像素集合,并构建以超像素为节点的图;然后融合超像素的协方差描述子、颜色信息、纹理信息、梯度信息以及边缘信息作为超像素的特征来度量超像素间的相似性,进而得到超像素的相似度矩阵;最后使用NJW算法对超像素图进行分割。大量的实验结果验证表明,改进的分割方法在分割精度上优于目前存在的无监督分割方法,并且在交互式分割的模式下,该方法可以准确分割出用户指定的目标。

关键词: 谱聚类, 图像分割, 相似度度量, 超像素, 协方差, NJW算法