Abstract
Synthetic aperture radar (SAR) has operational applications in crop mapping and monitoring in many countries due to the distinct backscatter signature at various stages of crop growth. Polarimetric analysis of SAR data from different satellites was used for information extraction from different types of scatters in imaged terrain. The scattering processes were analyzed through the received scatter matrix derived from the target decomposition of SAR data. Three decomposition techniques, namely Freeman–Durden, Cloude–Pottier and Touzi decomposition of the ALOS PALSAR-1 data, were used in this study to extract land use/cover information with a specific emphasis on agriculture. The decomposed output parameters from these techniques were classified with supervised classifier of support vector machine (SVM) using region of interest (ROI) selected land use/cover classes. An accuracy assessment for the classified output was carried out using the ROI. The Ramgarh village in Jaisalmer district of Rajasthan with the predominance of agricultural land, sand dunes and settlements was chosen as the study area. Freeman–Durden decomposition resulted in the highest overall accuracy of about 85% in the land use/cover classification among the three decomposition techniques adopted in the study. It was also observed that the accuracy of land use/cover mapping derived from Cloude–Pottier and Touzi decompositions improved with the use of eigenvalues in the SVM classification. Higher accuracies in the classification of agriculture land were noted with all the three decomposition techniques. The four parameters of Cloude–Pottier (H, A, α, β) and Touzi (α s, Φ s, ψ, τ) decompositions improved the classification accuracy for all the classes due to eigenvalues. The overall classification accuracy was above 88% for both the decomposition techniques with four parameters. The soil moisture values for agriculture land and sand dunes were validated through soil moisture maps generated using Oh 1992 and 2004 models.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.