Model-Based Deep Autoencoder Networks for Nonlinear Hyperspectral Unmixing
This work addresses robustness issues in unsupervised hyperspectral unmixing for remote sensing applications, representing an incremental improvement over existing autoencoder-based methods.
The paper tackles the problem of nonlinear hyperspectral unmixing by proposing a model-based autoencoder that incorporates prior information about the mixing process as a nonlinear fluctuation over a linear mixture, leading to improved performance in simulations with synthetic and real data.
Autoencoder (AEC) networks have recently emerged as a promising approach to perform unsupervised hyperspectral unmixing (HU) by associating the latent representations with the abundances, the decoder with the mixing model and the encoder with its inverse. AECs are especially appealing for nonlinear HU since they lead to unsupervised and model-free algorithms. However, existing approaches fail to explore the fact that the encoder should invert the mixing process, which might reduce their robustness. In this paper, we propose a model-based AEC for nonlinear HU by considering the mixing model a nonlinear fluctuation over a linear mixture. Differently from previous works, we show that this restriction naturally imposes a particular structure to both the encoder and to the decoder networks. This introduces prior information in the AEC without reducing the flexibility of the mixing model. Simulations with synthetic and real data indicate that the proposed strategy improves nonlinear HU.