 |
Center for Nonlinear Studies |
Hyperspectral imaging (HSI) is an imaging technology that provides fully registered spatial and high spectral resolution (radiance, reflectance, or emission) information of the scene in the field of view of the sensor. Hyperspectral remote sensing is currently undergoing a revolution with the appearance and blooming development of hyperspectral imaging sensors available across a number of platforms such as UAV, stand-off, and commercial/military airborne and space-borne systems. It is possible to capture information about a region of interest at high resolution in the spatial, spectral and temporal domains. Therefore it is of great interest the development of models and algorithms for hyperspectral image processing that fully exploit the information in all three domains. This presentation will describe ongoing research work in integrating spatial and spectral information for hyperspectral image processing. We present two perspectives on how to integrate spatial and spectral information. In the first one, we think of the image cube as a vector valued function where diffusion PDEs are used for image enhancement and scale-space representations that are used as preprocessing stages in hyperspectral image exploitation tasks. In the second one, we think of the image cube as a 3D tensor array with locality properties that can be used in improved hyperspectral unmixing. We illustrate potential benefits of these approaches with real data sets in solving classification, unmixing, and interest point extraction in hyperspectral image processing.
Host: Curt Canada