Superpixel-Based Autoencoder-Like Nonnegative Tensor Factorization for Hyperspectral Unmixing

Hyperspectral unmixing is significant for advancing remote sensing (RS) applications, aiming at extracting the spectra of pure materials (called endmembers) and obtaining their proportions (called fractional abundances) from an observed hyperspectral image (HSI). Nonnegative matrix factorization (NMF) is a popular technique for hyperspectral unmixing; however, it does not effectively preserve the spatial and spectral correlation of HSIs and fully characterize spectral variability. To overcome these limitations, we propose a novel superpixel-based autoencoder-like nonnegative tensor factorization (SANTF) model for hyperspectral unmixing. Specifically, drawing inspiration from the architecture of autoencoder, an autoencoder-like nonnegative tensor factorization (ANTF) model is constructed to directly project the hyperspectral data into abundance space. To further exploit the local spatial information, the superpixel strategy is incorporated into the ANTF framework, thereby building the SANTF model. Note that the superpixel cubes are jointly factorized instead of being represented as an average. Meanwhile, each superpixel cube generates one endmember matrix to capture spectral variability. Subsequently, a robust double weighted endmember (rDWE) constraint is designed to obtain the consensus endmember matrix adaptively. Moreover, the tensor nuclear norm (TNN) constraint is employed to enhance the low-rank characteristic of abundance tensor. The experimental results on both synthetic and real HSI datasets, compared with several unmixing methods, demonstrate that the proposed SANTF method can achieve superior unmixing performance.