Hyperspectral image classification via discriminative convolutional neural network with an improved triplet loss

作者：

Highlights：

• Introduce deep metric learning for HIS classification.

• Propose an improved triplet loss to achieve better performance.

• Design an appropriate network architecture.

• Improve the classification accuracy significantly.

摘要

•Introduce deep metric learning for HIS classification.•Propose an improved triplet loss to achieve better performance.•Design an appropriate network architecture.•Improve the classification accuracy significantly.

论文关键词：Hyper-spectral image classification,Convolutional neural network,Triplet loss,Discriminative learning,Metric learning

论文评审过程：Received 7 August 2019, Revised 12 July 2020, Accepted 31 October 2020, Available online 11 November 2020, Version of Record 30 January 2021.

论文官网地址：https://doi.org/10.1016/j.patcog.2020.107744