Spatial multiple materials detection in hyperspectral imagery

Nowadays the detection in space environment is definitely a hot spot in the whole world. Developing the technique of multiple materials detection makes great sense. In this paper, we propose an algorithm for spatial multiple materials detection in hyperspectral images. The detection problem is a semi-blind signal extraction problem. As the prior knowledge, the spectra of target materials are known in advance. The proposed detection algorithm, multiple materials detector (MMD), exploits spectral information exclusively to make decisions by considering that the type of each pixel contains the interesting materials or not, which is a semi-blind signal extraction method. With this method, after single time detection of a hyperspectral image, the multiple materials spectra input before could be exactly detected. Compared with classical detection methods, the proposed detection method has three superiorities. Firstly, it functions well when there are various kinds of interesting materials needing to be detected. Secondly, because of using regularization items the algorithm is robust in spectral variability and noise. Lastly, no matter how the interesting materials distribute in the hyperspectral image, it works steadily. Experimental results based on a set of hyperspectral images of Hubble Space Telescope prove the effectiveness of the MMD algorithm.