Soil moisture information extraction based on integration of active and passive remote sensing data

For improving the precision of soil moisture monitoring, a classifier based on integration of both active and passive remote sensing data and the Bayesian Networks for inversion of soil moisture was presented and tested in Heihe river basin, a semi-arid area in the north-west of China. In the algorithm the wavelet transform and IHS were combined to integrate TM3, TM4, TM5 and ASAR data. The method of maximum distance in local region was adopted as the fusion rule for prominent expression of the detailed information in the fusion image, and the spectral information of TM could be retained. Then the new R、G、B components in the fusion image and the TM6 were used as the input of the Bayesian network, and the outputs were five different categories corresponding to different levels of soil moisture values. The field measurement was carried out for validation of the method. A better result was acquired in vegetation coverage area, and the precision of classification could reach up to 76.1%, but ineffective in desert areas. So the method is applicable for reflecting the distribution of soil moisture in vegetation covered area.