Normalized Difference of Polinsar Coherence for Classification of Earth Covers

Abstract

The use of the interferometric SAR is extremely urgent for conditions of Siberia, nearly accessible and often covered by clouds. The polarimetric radar interferometry (PolInSAR) makes it possible to analyze the coherence of interferometric couples that provides additional information. This work makes an attempt to use a difference of polarimetric coherence value for the different types of classifications of the earth’s surface. As the processing method, it is proposed to use a widely known normalized difference vegetation index for interferometric coherence on different polarimetric combinations. It is established that two-dimensional brightness distribution of the coherence images on different polarimetric combinations is analogous with a similar distribution for multispectral data in the coordinates NIR- red. This distribution has angular dependence; therefore, it is necessary to use a procedure of the calculation of ratio-indexes. It is shown that the proposed approach improves the classification of earth covers.

Keywords: Interferometric, polarimetric radar interferometry, ratio-index, multispectral data