Hyperspectral multi-level image thresholding using qutrit genetic algorithm

作者：

Highlights：

• A Qutrit Genetic Algorithm is proposed for thresholding of hyperspectral images.

• A new quantum mutation procedure is implemented to bring diversity in the offsprings.

• Interactive Information method and BS-NET-Conv Network reduce the number of bands.

• Modified Otsu Criterion and Masi entropy are used as fitness functions.

• A quantum disaster operation prevents the population to get stuck into local optima.

摘要

•A Qutrit Genetic Algorithm is proposed for thresholding of hyperspectral images.•A new quantum mutation procedure is implemented to bring diversity in the offsprings.•Interactive Information method and BS-NET-Conv Network reduce the number of bands.•Modified Otsu Criterion and Masi entropy are used as fitness functions.•A quantum disaster operation prevents the population to get stuck into local optima.

论文关键词：Hyperspectral image thresholding,Quantum genetic algorithm,Quantum mutation operator,Multilevel quantum systems,Qutrit

论文评审过程：Received 16 July 2020, Revised 1 April 2021, Accepted 21 April 2021, Available online 29 April 2021, Version of Record 16 May 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.115107