Dynamical spectral unmixing of multitemporal hyperspectral images
This addresses the challenge of analyzing multitemporal hyperspectral data for applications like environmental monitoring, but it appears incremental as it builds on existing linear mixing models with a dynamical extension.
The paper tackled the problem of unmixing time series of hyperspectral images by proposing a dynamical model with latent variables for spectral signatures and fractional abundances, and demonstrated its performance on synthetic and real data.
In this paper, we consider the problem of unmixing a time series of hyperspectral images. We propose a dynamical model based on linear mixing processes at each time instant. The spectral signatures and fractional abundances of the pure materials in the scene are seen as latent variables, and assumed to follow a general dynamical structure. Based on a simplified version of this model, we derive an efficient spectral unmixing algorithm to estimate the latent variables by performing alternating minimizations. The performance of the proposed approach is demonstrated on synthetic and real multitemporal hyperspectral images.