Nonlocal graph theory based transductive learning for hyperspectral image classification

作者：

Highlights：

• The NLM similarity feature with full dimensionality of HSI data is exploited.

• A trunsductive variational model with nonlocal sparse graph expression is presented.

• A fast transductive alternating minimization iteration algorithm is designed to solve the classification problem.

• The classification results exhibit the excellent classification performance of the proposed method.

摘要

•The NLM similarity feature with full dimensionality of HSI data is exploited.•A trunsductive variational model with nonlocal sparse graph expression is presented.•A fast transductive alternating minimization iteration algorithm is designed to solve the classification problem.•The classification results exhibit the excellent classification performance of the proposed method.

论文关键词：Transductive learning,Nonlocal graph,Label propagation,Variational method,Alternating direction method of multipliers,Hyperspectral image classification

论文评审过程：Received 26 June 2020, Revised 25 November 2020, Accepted 26 March 2021, Available online 2 April 2021, Version of Record 12 April 2021.

论文官网地址：https://doi.org/10.1016/j.patcog.2021.107967