Clustering stability for automated color image segmentation

作者：

Highlights：

• We propose a clustering validation method (CS) to automatically segment images.

• We adapt CS to detect the best settings for color-texture feature extraction.

• Our method provides a score profile to compare alternative segmentation solutions.

• We test our procedure on textures, color natural images and 3D MRI data.

• Our method outperforms results obtained with other clustering validation indexes

摘要

•We propose a clustering validation method (CS) to automatically segment images.•We adapt CS to detect the best settings for color-texture feature extraction.•Our method provides a score profile to compare alternative segmentation solutions.•We test our procedure on textures, color natural images and 3D MRI data.•Our method outperforms results obtained with other clustering validation indexes

论文关键词：Clustering validation,Image segmentation,Clustering stability

论文评审过程：Received 28 October 2016, Revised 2 May 2017, Accepted 27 May 2017, Available online 31 May 2017, Version of Record 30 June 2017.

论文官网地址：https://doi.org/10.1016/j.eswa.2017.05.064