Hyperspectral image classification using Non-negative Tensor Factorization and 3D Convolutional Neural Networks

作者：

Highlights：

• HSI classification method is proposed using convolutional neural network.

• Non-negative Tensor Factorization methods are proposed to extract discriminant features.

• Morphological attribute filters are utilized for improving spatial context.

• Satisfying classification performance is achieved compared to other state-of-the art methods.

摘要

•HSI classification method is proposed using convolutional neural network.•Non-negative Tensor Factorization methods are proposed to extract discriminant features.•Morphological attribute filters are utilized for improving spatial context.•Satisfying classification performance is achieved compared to other state-of-the art methods.

论文关键词：Hyperspectral image classification,Non-negative Tensor Factorization (NTF),Convolutional Neural Network (CNN)

论文评审过程：Received 25 October 2018, Revised 5 March 2019, Accepted 7 May 2019, Available online 14 May 2019, Version of Record 16 May 2019.

论文官网地址：https://doi.org/10.1016/j.image.2019.05.004