HUPERSPECTRAL DATA ANALYSIS

Algorithms of linear spectral mixture analysis for hyperspectral images using base map

A. Yu. Denisova^ab, V. V. Myasnikov<sup>ab

a</sup> Samara State Aerospace University
^b Image Processing Systems Institute, Russian Academy of Sciences

Abstract: The authors propose two algorithms of linear spectral mixture analysis for hyperspectral images using base map. Base map data are used to refine coefficients of spectral mixture on the edges of map objects (in first algorithm) and acquire spectral signatures of small objects (in second algorithm), that does not occupy any pixel on input image entirely. The set of mixed signatures may be already known with undefined coefficients or unknown with extraction on one of the stages of the algorithm.

Keywords: hyperspectral images, spectral unmixing, hyperspectral analysis, subpixel selection, least squares method, base map.

Received: 29.03.2014