Binary Partition Tree construction from multiple features for image segmentation

作者：

Highlights：

• We propose a multi-feature BPT construction framework for image segmentation purpose.

• A collaborative strategy is proposed to establish a consensus among various metrics.

• It provides alternatives to the issue of metric construction from several (possibly non-comparable) features.

• Experiments highlight its versatility to build BPT from multiple images or metrics.

摘要

•We propose a multi-feature BPT construction framework for image segmentation purpose.•A collaborative strategy is proposed to establish a consensus among various metrics.•It provides alternatives to the issue of metric construction from several (possibly non-comparable) features.•Experiments highlight its versatility to build BPT from multiple images or metrics.

论文关键词：Binary Partition Tree,Morphological hierarchies,Multiple features,Graph-based image processing,Image segmentation

论文评审过程：Received 4 September 2017, Revised 24 May 2018, Accepted 1 July 2018, Available online 24 July 2018, Version of Record 24 July 2018.

论文官网地址：https://doi.org/10.1016/j.patcog.2018.07.003