Novel hyperbolic clustering-based band hierarchy (HCBH) for effective unsupervised band selection of hyperspectral images
作者：
Highlights：
• A novel hyperbolic clustering-based band hierarchy for effective unsupervised band selection.
• A novel hyperbolic distance for measuring the geometrical relationship between different bands.
• A novel cluster-centre ranking strategy for selecting significant band from each cluster.
• Validated robustness on several public available hyperspectral datasets.
摘要
•A novel hyperbolic clustering-based band hierarchy for effective unsupervised band selection.•A novel hyperbolic distance for measuring the geometrical relationship between different bands.•A novel cluster-centre ranking strategy for selecting significant band from each cluster.•Validated robustness on several public available hyperspectral datasets.
论文关键词：Hyperspectral image,Unsupervised band selection,Hyperbolic space clustering,Hierarchical clustering
论文评审过程：Received 19 March 2022, Revised 3 May 2022, Accepted 11 May 2022, Available online 13 May 2022, Version of Record 16 May 2022.
论文官网地址：https://doi.org/10.1016/j.patcog.2022.108788
