 
Summary: Unmixing and target recognition in hyperspectral1
images2
Amir Z. Averbuch Valery Zheludev Michael V. Zheludev
School of Computer Science
Tel Aviv University
Tel Aviv 69978, Israel
3
Abstract4
We present two new linear algorithms that perform unmixing in hyperspectral5
images and then recognize their targets whose spectral signatures are given. The6
first algorithm is based on the ordered topology of spectral signatures. The second7
algorithm is based on a linear decomposition in each pixel's neighborhood. The sought8
after target can occupy sub or above pixel. These algorithms combine ideas from9
algebra and probability theories as well as statistical data mining. Experimental results10
demonstrate their robustness.11
1 Introduction12
1.1 Data representation and extraction of spectral information13
We assume that an hyperspectral signature of a sought after material is given. In many14
applications, according to (Winter, 1999), a fundamental processing task is to automatically15
identify pixels whose spectra coincide with the given spectral shape (signature). This prob16
