Hyperspectral remote sensing image classification based on tighter random projection with minimal intra-class variance algorithm

作者：

Highlights：

• A significant improvement on Gaussian dimensional bounds for RP is proposed with detailed proved.

• The number of spectral vectors of the proposed algorithm is larger than that of the traditional RP.

• Considering the class separability, the TRP-MIV matrix with sample assistance provides a promising avenue for dimensionality reduction of hyperspectral remote sensing image.

• It is the first application of the TRP-MIV algorithm for hyperspectral remote sensing image classification.

摘要

•A significant improvement on Gaussian dimensional bounds for RP is proposed with detailed proved.•The number of spectral vectors of the proposed algorithm is larger than that of the traditional RP.•Considering the class separability, the TRP-MIV matrix with sample assistance provides a promising avenue for dimensionality reduction of hyperspectral remote sensing image.•It is the first application of the TRP-MIV algorithm for hyperspectral remote sensing image classification.

论文关键词：Random projection,Dimensionality reduction,Image size,Minimum distance classifier,Hyperspectral remote sensing image classification

论文评审过程：Received 8 April 2020, Revised 9 August 2020, Accepted 6 September 2020, Available online 11 September 2020, Version of Record 19 September 2020.

论文官网地址：https://doi.org/10.1016/j.patcog.2020.107635