Nonlinear unmixing of hyperspectral images using a semiparametric model and spatial regularization

2013/10/31 by Jie Chen, Chen, Jie, Cédric Richard +4
Earth and Planetary Sciences · Engineering · Mathematics · #Advanced Image Fusion Techniques #FOS: Computer and information sciences #Machine Learning (stat.ML) #Remote Sensing and Land Use #Remote-Sensing Image Classification #stat.ML

paper · pdf · doi:10.48550/arxiv.1310.8612

5 pages, 1 figure, submitted to ICASSP 2014

arxiv created 2013/10/31 · openalex publication_date 2013/10/31 · arxiv updated 2013/11/01 · openalex created_date 2019/06/27 · openalex updated_date 2026/07/28

Abstract

Incorporating spatial information into hyperspectral unmixing procedures has been shown to have positive effects, due to the inherent spatial-spectral duality in hyperspectral scenes. Current research works that consider spatial information are mainly focused on the linear mixing model. In this paper, we investigate a variational approach to incorporating spatial correlation into a nonlinear unmixing procedure. A nonlinear algorithm operating in reproducing kernel Hilbert spaces, associated with an ℓ₁ local variation norm as the spatial regularizer, is derived. Experimental results, with both synthetic and real data, illustrate the effectiveness of the proposed scheme.

Related

• Using Spatial Correlation in Semi-Supervised Hyperspectral Unmixing under Polynomial Post-nonlinear Mixing Model
• Total variation regularized non-negative matrix factorization for smooth hyperspectral unmixing
• Bayesian nonlinear hyperspectral unmixing with spatial residual component analysis
• Fast Hyperspectral Unmixing in Presence of Nonlinearity or Mismodelling Effects
• Towards Robust Hyperspectral Unmixing: Mixed Noise Modeling and Image-Domain Regularization