M³FuNet: An Unsupervised Multivariate Feature Fusion Network for Hyperspectral Image Classification

Hyperspectral image (HSI) spectral-spatial joint feature (FE) extraction methods generally suffer from low feature retention and weak spatial–spectral dependence, which will lead to single-class feature confrontation (SCFC). To solve this problem, an unsupervised multivariate feature fusion network (M3FuNet) is developed in this article. In M3FuNet, multiscale supervector matrix correction (MSMC) and multiscale random convolution dispersion (MRCD) are used as the spectral and spatial feature extraction method, and the feature retention of spectral and spatial features is improved to achieve feature calibration by feature fusion and decision fusion, called “multivariate feature fusion.” The MSMC is employed to correct the supervector matrix to reduce the intraclass variance in superpixel homogeneous regions and overcome the phenomenon of supervector block drift (SvBD). The MRCD uses random convolution and Gaussian smoothing to extract deep spatial features. Because of the similar feature representation ability of the MSMC and MRCD, the obtained spectral-spatial joint features have high feature retention and strong spectral-spatial dependence. Finally, this M3FuNet is used for realizing the classification of HSI. Three common HSI datasets are used to validate the effectiveness of the M3FuNet. The experiment results show that the M3FuNet has a superior performance compared with several state-of-the-art (SOTA) HSI classification methods. The code of the proposed M3FuNet is available at
https://github.com/aichou233/M3FuNet
.