Details

Autor(en) / Beteiligte

• Sun, Le
• Zhao, Guangrui
• Zheng, Yuhui
• Wu, Zebin

Titel

Spectral-Spatial Feature Tokenization Transformer for Hyperspectral Image Classification

Ist Teil von

• IEEE transactions on geoscience and remote sensing, 2022, Vol.60, p.1-14

Ort / Verlag

New York: IEEE

Erscheinungsjahr

2022

Quelle

IEEE Electronic Library (IEL)

Beschreibungen/Notizen

• 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's number increases, computational costs rise significantly. The transformer framework can represent high-level semantic features well. In this article, a spectral-spatial feature tokenization transformer (SSFTT) method is proposed to capture spectral-spatial features and high-level semantic features. First, a spectral-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.