Rocks/Minerals Information Extraction from EO-1 Hyperion Data Base on SVM

Hyperspectral remote sense image have been used successfully for mineral exploration. The high dimensionality of such images arise various problems like curse of dimensionality and large hypothesis space. In this paper we make an approach to the application of support vector machine theory in rocks/minerals information extraction from EO-1 Hyperion data. The first, we present a feature extraction method based on Automatic Subspace Partition (ASP). The hyperspectral data bands are firstly partitioned different subspaces base on neighboring correlation of bands and extracted spectral feature of different subspace. Then we employed the support vector machine (SVM) classifier for classification and rocks/minerals information extraction. Two Hyperion images of the BeiYa in the northwest of YunNan was acquired and evaluated for alteration zone mapping. The results show that the alteration zones in the study area can be identified from Hyperion data very efficiently. The mineralogical and litho logic information extracted from Hyperion data is largely consistent with the geological map and previous research results.