Self-paced learning-assisted regularization reconstruction method with data-adaptive prior for electrical capacitance tomography
作者：
Highlights：
• A new optimization problem is crafted to model the inverse imaging problem.
• A new optimizer is proposed to solve the built optimization model.
• The fusion of domain knowledge and dada-adaptive knowledge is achieved.
• The confluence of supervised learning and imaging problem is realized.
• The performance gain of the proposed imaging technique is verified.
摘要
•A new optimization problem is crafted to model the inverse imaging problem.•A new optimizer is proposed to solve the built optimization model.•The fusion of domain knowledge and dada-adaptive knowledge is achieved.•The confluence of supervised learning and imaging problem is realized.•The performance gain of the proposed imaging technique is verified.
论文关键词：Computational imaging,Random forest,Self-paced learning,Inverse problem,Electrical capacitance tomography
论文评审过程：Received 12 July 2020, Revised 21 October 2020, Accepted 24 November 2021, Available online 29 November 2021, Version of Record 6 December 2021.
论文官网地址：https://doi.org/10.1016/j.eswa.2021.116296
