SGT: A Generalized Processing Model for 1-D Remote Sensing Signal Classification

This letter proposes a generalized feature extraction framework for 1-D remote sensing data. This approach streamlines the processing for extracting features by eliminating the need for some preprocessing, such as data normalization, data filtering, and spectrogram generation, which explicitly encode domain-specific knowledge of the tasks. The main component of the new framework, called shifted-grad Transformer (SGT), includes the shift module, grad module, smooth module, raw embedding module, Transformer encoder module, and additional essential module. Extensive experiments on datasets, such as hyperspectral image (HSI) data, magnetic signal data, and other 1-D data, have demonstrated that the SGT performs significantly better than the existing methods and provides a new solution to the 1-D data processing problem. Our training code and data are available at
https://github.com/wfnian/SGT
.