A 3-dimensional polynomial model for geometric corrections of airborne SAR images

Abstract

Terrain undulations have significant influence on the geometric and radiometric quality of high-resolution airborne synthetic aperture radar images. The correction of these effects becomes indispensable before image analyzing. The paper presents a new approach to the geometric correction under unknown parameters of the SAR imaging geometry. Using a three-dimensional (3D) polynomial model, the image of the testing area is geo-corrected. The selection of polynomial order is emphasized. The best appropriate polynomial order for testing area is selected after testing nine different polynomial orders which are 1-1-0, 1-1-1, 2-2-0, 2-2-1, 2-2-2, 3-3-0, 3-3-1, 3-3-2, 3-3-3. The resulting image is a fully terrain corrected one and could be used for real-time monitoring and evaluating emergencies (e.g., hurricanes or earthquakes).