Classification of imbalanced hyperspectral images using SMOTE-based deep learning methods

作者：

Highlights：

• The imbalanced classification problem is solved.

• A new hybrid SMOTE- based is proposed.

• The best model is Smote Balanced 5-CV multiclass classification.

摘要

•The imbalanced classification problem is solved.•A new hybrid SMOTE- based is proposed.•The best model is Smote Balanced 5-CV multiclass classification.

论文关键词：Hyperspectral imaging,Deep learning,Image classification,SMOTE,Imbalanced dataset,Imbalanced classification

论文评审过程：Received 17 March 2021, Revised 27 March 2021, Accepted 31 March 2021, Available online 6 April 2021, Version of Record 22 April 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.114986