A new method for building adaptive Bayesian trees and its application in color image segmentation
作者：
Highlights：
• A novel clustering method applied to color image segmentation is proposed.
• A Bayesian framework is used to detect modes of the image features distribution.
• Clusters are built based on a hierarchy of directed trees structures.
• The clusters are used to perform color image segmentation.
• The method potentially can be reliable based on experiments and method comparisons.
摘要
•A novel clustering method applied to color image segmentation is proposed.•A Bayesian framework is used to detect modes of the image features distribution.•Clusters are built based on a hierarchy of directed trees structures.•The clusters are used to perform color image segmentation.•The method potentially can be reliable based on experiments and method comparisons.
论文关键词：Clustering,Color image segmentation,Directed trees,Bayesian decision theory
论文评审过程：Received 26 April 2017, Revised 8 December 2017, Accepted 24 December 2017, Available online 25 December 2017, Version of Record 13 January 2018.
论文官网地址：https://doi.org/10.1016/j.eswa.2017.12.045
