Two-stage image segmentation based on nonconvex ℓ2−ℓp approximation and thresholding
作者：
Highlights：
• We propose a new nonconvex l2−lp approximation of the Mumford–Shah model for image segmentation.
• Our model adopts a two-stage image segmentation strategy and we use a closed-form p-shrinkage operator to deal with the lp quasi-norm subproblem in the first stage.
• We utilize the K-means method to select the thresholds automatically and get good segmentation results both qualitatively and quantitatively.
摘要
•We propose a new nonconvex l2−lp approximation of the Mumford–Shah model for image segmentation.•Our model adopts a two-stage image segmentation strategy and we use a closed-form p-shrinkage operator to deal with the lp quasi-norm subproblem in the first stage.•We utilize the K-means method to select the thresholds automatically and get good segmentation results both qualitatively and quantitatively.
论文关键词：Image segmentation,Two-stage strategy,Split–Bregman,Nonconvex approximation
论文评审过程：Received 27 May 2019, Revised 24 June 2020, Accepted 7 March 2021, Available online 19 March 2021, Version of Record 19 March 2021.
论文官网地址：https://doi.org/10.1016/j.amc.2021.126168
