Envisat ASAR Data Processing And Its Application In Forest And Agriculture Monitoring

Synthetic Aperture Radar (SAR) is anticipated to be the dominant high-resolution remote sensing data source for agricultural applications in tropical and subtropical regions due to its all-weather and all-time imaging. Envisat-1 ASAR is the most advanced satellite radar-imaging instrument, its important new capabilities include beam steering for acquiring images with different incidence angles, duel polarization and wide swath coverage. So it has good application prospect. Based on the mechanism of microwave remote sensing in agriculture and forest, this article discusses Envisat-1 ASAR data preprocessing method and the potential abilities of its application. The main content of this paper is as follows. 1. The method research of Envisat-1 ASAR data preprocessing includes radiometric calibration, geometrical correction and speckle reduction. The quality of preprocessing is important for the application's reliability and accuracy. The preprocessing study is based on existing commercial software. DEM is necessary in the geometrical correction of non-flat area. SAR simulation is helpful to the selection of GCPs in hilled area. For high accuracy of less than one pixel, auto-registration technique was adopted. Multi-channel filtering can be carried out in an optimal way, which minimizes speckle while preserving the radiometry and spatial resolution of the individual channels. 2. Comparison of the different classification methods. SAR data classification is quite different from optical images because of the special imaging mechanism. Three methods are compared, including traditional supervised classification, expert classification and object oriented classification. The object oriented classification is the best for SAR images. 3. Research of Envisat-1 ASAR data application in agriculture. Multi-temporal ASAR data are used to identify rice and banana in the regions of Zhagnzhou and Fuzhou. The backscattering characteristics of rice change sharply with the temporal. The early stage of growth is important for rice classification in the research site. However the backscattering characteristics of banana change little with season, so the temporal selection is less important in banana identification. Principle component analysis was adopted for the classification for rice and banana, PC1 is used in banana and PC2 is used in rice. Duel polarization information is good for rice and banana classification. 4. Research of Envisat-1 ASAR data application in littoral shelter forest. The backscattering characteristics of littoral shelter forest change little with season, the polarization information is more important than the temporal. 5. Researching of different SAR data application in forest and agriculture classification. Envisat-1 ASAR,ERS-1 SAR and JERS-1 SAR all can be used in crop identification if the temporal selection are fit. VV and VH polarization are better for forest classification than HH.