A goal-driven unsupervised image segmentation method combining graph-based processing and Markov random fields

作者：

Highlights：

• An unsupervised and graph-based image segmentation method is proposed.

• It is equipped with a goal-oriented component, favoring subsequent analysis.

• It is driven by the intrinsic properties of the data and by the goal to be achieved.

• Real applications are examined in the biomedical and remote sensing framework.

• Experiments within different domains highlight the method’s robustness and adaptation capability.

摘要

•An unsupervised and graph-based image segmentation method is proposed.•It is equipped with a goal-oriented component, favoring subsequent analysis.•It is driven by the intrinsic properties of the data and by the goal to be achieved.•Real applications are examined in the biomedical and remote sensing framework.•Experiments within different domains highlight the method’s robustness and adaptation capability.

论文关键词：Graph signal processing,Segmentation,Markovian modeling,Parametric model estimation,Pattern recognition,Synthetic aperture radar,Magnetic resonance imagery

论文评审过程：Received 19 December 2021, Revised 4 August 2022, Accepted 27 September 2022, Available online 30 September 2022, Version of Record 18 October 2022.

论文官网地址：https://doi.org/10.1016/j.patcog.2022.109082