Graph Guided Transformer: An Image-Based Global Learning Framework for Hyperspectral Image Classification
Chengzhong Shi, Qiming Liao, Xinping Li, Lin Zhao, Wen Li
Abstract
Hyperspectral image (HSI) classification methods often follow an approach of patch-based learning framework. Recently, an image-based global deep learning framework has gained increasing attention for HSI classification tasks due to faster inference speed. However, such a framework exhibits deteriorated performance in modeling features on the region level while balancing local spatial structure information. In this letter, we propose a global learning method that includes graph-guided transformer (G <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> T) as the core tool. First, we extract pixel level features by convolution block and obtain an undirected graph by segmentation on superpixel scales for an input HSI. Then, to model global and local correlations among nodes of superpixels, a mechanism of graph-guided self-attention (G <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> SA) is developed and implemented. Finally, pixel level features integrated with superpixel features at regional level are used to generate classification results for the HSI. Experimental results demonstrate that the method of G <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> T outperforms state-of-the-art methods in classification accuracy and inference speed, in particular in the case of limited labeled sample. The source code for this work will be available at https://github.com/zhaolin6/G2T.