An Approach to Spectral Discrimination of Rocks Using ASAI and Rough Sets

Field data of outcrop spectrums provide important basis for modeling of hyper-spectral remote sensing aiming at mineral prospecting. We make an approach to the application of rough set theory in spectral discrimination of rocks. We build a decision table with an adequate number of samples (outcrops) of known rock type (the universe), of which the conditional attributes are discretized 'area spectrum absorption indexes' (ASAI) corresponding to wavelength intervals, and the decision attribute is rock type. We search to obtain the exhaustive set of reducts of the table, each of which will serve as a variable number of deduction rules. Suppose we have n (usually a very big number) rules in total and there are m types of rocks in our universe, for any unknown sample, we judge its rock type by each of those rules. An unknown sample may be recognized as a different type by different rules because it is outside our universe, and we accept the most frequent judgment result and ignore the other m-1 types of results. Our ASAI is an improvement upon the traditional spectrum absorption index (SAI), better applicable to field spectrums: given a spectrum curve and a wavelength interval, we take the average reflectance within the interval as a base line and let ASAI=a below/(a above+a below), where a below and a above stand for total areas, bounded by the curve, the base line and the borders of the intervalbelow and above the base line respectively. With the equipments of FieldSpectr Fr (made by ASD Co., US), we collected data from Baiya gold deposit, Yunnan, and applied the above method to discriminate altered rocks as an experiment. The results show satisfactory performance of the method.

Hyperspectral remote sensing images terrain classification in DCT SRDA subspace

Hyperspectral remote sensing images terrain classification faces the problems of high data dimensionality and lack of labeled training data, resulting in unsatisfied terrain classification efficiency. The feature extraction is required before terrain classification for preserving discriminative information and reducing data dimensionality. A hyperspectral remote sensing images feature extraction method, i.e., discrete cosine transform （DCT） spectral regression discriminant analysis （SRDA） subspace method, was presented to solve the above problems. The proposed DCT SRDA subspace method firstly takes DCT in the original spectral space and gets the DCT coefficients of each pixel spectral curve; secondly performs SRDA in the DCT coefficients space and obtains the DCT SRDA subspace. Minimum distance classifier was designed in the resulting DCT SRDA subspace to evaluate the feature extraction performance. Experiments for two real airborne visible/infrared imaging spectrometer （AVIRIS） hyperspectral images show that, comparing with spectral LDA subspace method, the proposed DCT SRDA subspace method can improve terrain classification efficiency.