Unseen Feature Extraction: Spatial Mapping Expansion With Spectral Compression Network for Hyperspectral Image Classification

Authors:
Publisher website
Google Scholar
Abstract
Hyperspectral image classification (HSIC) models have made remarkable progress in the last decade. Nevertheless, the downsized mapping in the convolutional neural network (CNN) and down-sampled mechanism in the transformer-based approach amplify the loss of hidden knowledge in the subpixel that encompasses crucial yet unseen features within a single pixel. Considering this aspect, the mentioned popular solutions for HSIC contradict the inherent characteristic of hyperspectral data. To address this issue, we rethink the size factor in CNN and propose a novel spatial mapping expansion with spectral compression network (SMESC) for HSIC. Specifically, the SMESC builds a mapping expansion network to mine unseen information in subpixels with enlarged feature maps. A channel modulation residual block is developed to compress spectral redundancy and promote salient channels with modulation information. Moreover, we design a multiple-size training strategy to substitute the traditional multiple FE branches and improve the model adaptation to the different sizes of the testing samples. The extensive experimental results and analysis of four hyperspectral image datasets demonstrate the superiority of the proposed architecture compared to other advanced HSIC methods. Our code will be released at https://github.com/ Chirsycy/SMESC.
Funding Information
  • National Natural Science Foundation of China (61971082, 42271355, 42101350)
  • Science Foundation of Liaoning Province through the Surface Project (LJKZ0065)

This publication has 44 references indexed in Scilit: