An adaptive differential evolution algorithm to optimal multi-level thresholding for MRI brain image segmentation

作者：

Highlights：

• The ALDE algorithm works better for MRI image analysis than DE variants.

• The ALDE algorithm is less sensitive to increasing number of thresholds.

• The ALDE algorithm is fast enough for real-world image analysis applications.

• The ALDE algorithm can efficiently balance between exploration and exploitation.

摘要

•The ALDE algorithm works better for MRI image analysis than DE variants.•The ALDE algorithm is less sensitive to increasing number of thresholds.•The ALDE algorithm is fast enough for real-world image analysis applications.•The ALDE algorithm can efficiently balance between exploration and exploitation.

论文关键词：Image segmentation,Differential evolution,Optimal thresholding,Lévy distribution,Cauchy distribution,Cotes’ Spiral

论文评审过程：Received 21 February 2019, Revised 12 July 2019, Accepted 15 July 2019, Available online 16 July 2019, Version of Record 22 July 2019.

论文官网地址：https://doi.org/10.1016/j.eswa.2019.07.037