Journal
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
Volume 60, Issue -, Pages -Publisher
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TGRS.2022.3144158
Keywords
Feature extraction; Transformers; Convolution; Semantics; Principal component analysis; Data mining; Convolutional neural networks; Convolutional neural networks (CNNs); hyperspectral image (HSI) classification; semantic features; spectral-spatial tokenization; transformer
Categories
Funding
- National Natural Science Foundation of China [61971233, 62076137, U20B2065, U20B2061]
- Natural Science Foundation of Jiangsu Province [BK 20211539]
- Henan Key Laboratory of Food Safety Data Intelligence [KF2020ZD01]
- Postgraduate Research & Practice Innovation Program of Jiangsu Province [KYCX21_1004]
Ask authors/readers for more resources
In this article, the spectral-spatial feature tokenization transformer (SSFTT) method is proposed to capture spectral-spatial and high-level semantic features. Experimental analysis confirms that this method outperforms other deep learning methods in terms of computation time and classification performance.
In hyperspectral image (HSI) classification, each pixel sample is assigned to a land-cover category. In the recent past, convolutional neural network (CNN)-based HSI classification methods have greatly improved performance due to their superior ability to represent features. However, these methods have limited ability to obtain deep semantic features, and as the layer & x2019;s number increases, computational costs rise significantly. The transformer framework can represent high-level semantic features well. In this article, a spectral & x2013;spatial feature tokenization transformer (SSFTT) method is proposed to capture spectral & x2013;spatial features and high-level semantic features. First, a spectral & x2013;spatial feature extraction module is built to extract low-level features. This module is composed of a 3-D convolution layer and a 2-D convolution layer, which are used to extract the shallow spectral and spatial features. Second, a Gaussian weighted feature tokenizer is introduced for features transformation. Third, the transformed features are input into the transformer encoder module for feature representation and learning. Finally, a linear layer is used to identify the first learnable token to obtain the sample label. Using three standard datasets, experimental analysis confirms that the computation time is less than other deep learning methods and the performance of the classification outperforms several current state-of-the-art methods. The code of this work is available at https://github.com/zgr6010/HSI_SSFTT for the sake of reproducibility.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available