Data acquired from multi-channel sensors is a highly valuable asset to interpret the environment for a variety of remote sensing applications. However, low spatial resolution is a critical limitation for these sensors and the constituent materials of a scene can be mixed in different fractions due to their spatial interactions. Spectral unmixing is a technique that allows us to obtain these material spectral signatures with their fractions from data. In this paper, we propose a novel hyperspectral unmixing scheme, called EndNet, that is based on a two-staged autoencoder. This well-known structure is completely enhanced by leveraging previous assumptions introduced for hyperspectral unmixing domain. Also, we make critical contributions to both architecture and optimization step to achieve state-of-the-art performance for hyperspectral data. To demonstrate the superiority of our method, we conduct extensive experiments on several datasets. The obtained results confirm that our method considerably improves the performance compared to the state-of-the-art techniques in literature.
View on arXiv