Illumination invariant hyperspectral image unmixing based on a digital surface model
This addresses the problem of inaccurate unmixing in shaded areas for remote sensing applications, representing an incremental improvement by integrating physical explanations into the framework.
The paper tackled spectral variability in hyperspectral image unmixing caused by variable illuminations and shadows by proposing an illumination invariant spectral unmixing (IISU) model that uses radiance data and a LiDAR-derived digital surface model. The result showed that IISU estimated more accurate abundances and shadow compensated reflectance than state-of-the-art models, especially in shaded pixels where other models did not work well.
Although many spectral unmixing models have been developed to address spectral variability caused by variable incident illuminations, the mechanism of the spectral variability is still unclear. This paper proposes an unmixing model, named illumination invariant spectral unmixing (IISU). IISU makes the first attempt to use the radiance hyperspectral data and a LiDAR-derived digital surface model (DSM) in order to physically explain variable illuminations and shadows in the unmixing framework. Incident angles, sky factors, visibility from the sun derived from the LiDAR-derived DSM support the explicit explanation of endmember variability in the unmixing process from radiance perspective. The proposed model was efficiently solved by a straightforward optimization procedure. The unmixing results showed that the other state-of-the-art unmixing models did not work well especially in the shaded pixels. On the other hand, the proposed model estimated more accurate abundances and shadow compensated reflectance than the existing models.