Unassisted thresholding based on multi-objective evolutionary algorithms

作者：

Highlights：

• Αn Unassisted Thresholding (UTH) is presented to perform image segmentation while simultaneously determinating the optimal number of thresholds and their values.

• UTH is implemented with NSGA III, MOPSO, and PESA II to evaluate its performance.

• A particle with variable length is used to encode threshold values on the population.

摘要

•Αn Unassisted Thresholding (UTH) is presented to perform image segmentation while simultaneously determinating the optimal number of thresholds and their values.•UTH is implemented with NSGA III, MOPSO, and PESA II to evaluate its performance.•A particle with variable length is used to encode threshold values on the population.

论文关键词：Multi-objective optimization,Evolutionary algorithms,NSGA-III,Cross entropy,Unassisted thresholding

论文评审过程：Received 6 December 2017, Revised 20 June 2018, Accepted 28 June 2018, Available online 6 July 2018, Version of Record 10 September 2018.

论文官网地址：https://doi.org/10.1016/j.knosys.2018.06.028