Saliency detection via multiple-morphological and superpixel based fast fuzzy C-mean clustering network

作者：

Highlights：

• The proposed technique produces the robust edge feature gradient image.

• A probability-based subtraction technique is applied to remove the background cluster.

• The saliency maps are fused by using consistency based fusion technique.

• We have carried out a comprehensive experiment on nine different data sets.

摘要

•The proposed technique produces the robust edge feature gradient image.•A probability-based subtraction technique is applied to remove the background cluster.•The saliency maps are fused by using consistency based fusion technique.•We have carried out a comprehensive experiment on nine different data sets.

论文关键词：Gradient image,Superpixel image,Background subtraction,Clustering,Saliency map

论文评审过程：Received 22 September 2019, Revised 20 May 2020, Accepted 10 June 2020, Available online 26 June 2020, Version of Record 13 July 2020.

论文官网地址：https://doi.org/10.1016/j.eswa.2020.113654