Content Information Extraction Methods Study Of Power Transmission Corridor Area Based On SAR Image

Soil moisture is the material basis of vegetation growth and development,and it can affect the vegetation distribution in different regions. Soil moisture plays an important role in forecasting landslides, debris flow, surface subsidence and other natural disasters. Although soil moisture inversion in flat area has been able to achieve high accuracy, it is still a problem to invert on mountainous.Compared with optical technology, microwave remote sensing could work all-day in all- weather, and there is no influence with light and cloud rain weather. Since the 20 th century, synthetic aperture radar technology had get great progress, Now multi-band, multi-polarization, multi-angle of incidence SAR data could obtain successfully. Compared with single-polarization, single-band SAR image data, multi-polarization data contains more soil moisture information, and it could make soil moisture retrieval more accurate, soil moisture retrieval model simpler.In this paper, the author extract soil moisture though the method of combining the theory with practice. June 2015, measured backscatter coefficient of soil with different content or different submerged. AIEM simulation data and experimental data been used to establish a bare soil moisture inversion model, and use backscatter coefficient of different submerged to establish a semi- flooded proportion inversion model. Different measurement data with different incidence angle is used to test the accuracy of experience model, validation results show that the models can invert soil moisture and the proportion of semi- submerged accuracy. In this paper, as the existence of vegetation area, combine water-cloud model and bare soil moisture inversion model establish vegetation cover soil moisture inversion algorithm.This study using the method extract soil moisture of Luding Radarsat-2 data, but there is a large undulating terrain, vegetation cover of the region. Radar system is a strabismus imaging system, the incident angel of Radarsat-2 data of this paper is 42.5°. Due to the two features above, there is a large shade area in SAR image. High-resolution DEM data extracted from InSAR technology with 90- meter resolution DEM data, use elevation correction way to remove shadow area of SAR image.The paper firstly use multi- view, filter, radiometric correction, geometric correction method, elevation correction to deal with HH/VV polarization SAR images data. Then using SVM method classify with post-processing VV polarization image, to remove water and shade areas of SAR image. Bare soil moisture inversion model is used bare area, vegetation cover soil moisture inversion model is used vegetation area. Finally, analyze and verify the water content inversion result of the transmission corridor area.