Abstract
Abstract. Due to the limited spatial resolution of remote hyperspectral sensors, pixels are usually highly mixed in the hyperspectral images. Endmember extraction refers to the process identifying the pure endmember signatures from the mixture, which is an important step towards the utilization of hyperspectral data. Nonnegative matrix factorization (NMF) is a widely used method of endmember extraction due to its effectiveness and convenience. While most NMF-based methods have single-layer structures, which may have difficulties in effectively learning the structures of highly mixed and complex data. On the other hand, multilayer algorithms have shown great advantages in learning data features and been widely studied in many fields. In this paper, we presented a L1 sparsityconstrained multilayer NMF method for endmember extraction of highly mixed data. Firstly, the multilayer NMF structure was obtained by unfolding NMF into a certain number of layers. In each layer, the abundance matrix was decomposed into the endmember matrix and abundance matrix of the next layer. Besides, to improve the performance of NMF, we incorporated sparsity constraints to the multilayer NMF model by adding a L1 regularizer of the abundance matrix to each layer. At last, a layer-wise optimization method based on NeNMF was proposed to train the multilayer NMF structure. Experiments were conducted on both synthetic data and real data. The results demonstrate that our proposed algorithm can achieve better results than several state-of-art approaches.
Highlights
Hyperspectral remote sensing images generally contain abundant spatial and spectral information of the covered areas, which can be useful in the applications of earth monitoring, land cover classification, mineral exploration, military surveillance, etc. (Tong et al, 2016)
Before utilizing the hyperspectral images, we need to decompose these mixed pixels into a set of endmembers and their corresponding abundances first
Several experiments are conducted on both synthetic data and real data to compare our method with several existing state-of-art approaches, namely the vertex component analysis method (VCA-fully constrained least squares method (FCLS)), the projected gradient Nonnegative matrix factorization (NMF) method (PGNMF) (Lin, 2007), the minimum volume constrained NMF method (MVC-NMF) (Miao et al, 2007), and the multilayer NMF method (MLNMF) (Rajabi et al, 2015)
Summary
Hyperspectral remote sensing images generally contain abundant spatial and spectral information of the covered areas, which can be useful in the applications of earth monitoring, land cover classification, mineral exploration, military surveillance, etc. (Tong et al, 2016). Due to the long observation distance and the low spatial resolution of hyperspectral sensors, pixels in the acquired hyperspectral images are usually a mixture of several ground cover spectrum. These pure spectra are known as endmembers and their proportions in the pixels are called the corresponding abundance fractions (Zhu et al, 2014). Before utilizing the hyperspectral images, we need to decompose these mixed pixels into a set of endmembers and their corresponding abundances first. This process is called hyperspectral unmixing, which contains two procedures: endmember extraction and abundance estimation (Miao et al, 2007)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callMore From: The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
