Georeferencing on Synthetic Aperture Radar Imagery

Abstract

ABSTRACT:Due to the SAR 1 geometry imaging, SAR images include geometric distortions that would be erroneous image information and the images should be geometrically calibrated. As the radar systems are side looking, geometric distortion such as shadow, foreshortening and layover are occurred. To compensate these geometric distortions, information about sensor position, imaging geometry and target altitude from ellipsoid should be available. In this paper, a method for geometric calibration of SAR images is proposed. The method uses Range-Doppler equations. In this method, for the image georeferencing, the DEM 2 of SRTM with 30m pixel size is used and also exact ephemeris data of the sensor is required. In the algorithm proposed in this paper, first digital elevation model transmit to range and azimuth direction. By applying this process, errors caused by topography such as foreshortening and layover are removed in the transferred DEM. Then, the position of the corners on original image is found base on the transferred DEM. Next, original image registered to transfer DEM by 8 parameters projective transformation. The output is the georeferenced image that its geometric distortions are removed. The advantage of the method described in this article is that it does not require any control point as well as the need to attitude and rotational parameters of the sensor. Since the ground range resolution of used images are about 30m, the geocoded images using the method described in this paper have an accuracy about 20m (subpixel) in planimetry and about 30m in altimetry.

Automatic refinement of ephemeris‐derived range‐Doppler model through the least‐squares matching of Sentinel‐1 Single‐Look Complex images and DEM in mountainous areas

“…Esmaeilzadeh and Amini by mapping the DEM into the SAR image space via the RD model increased the geometric similarity of these two datasets to facilitate the recognition of the correspondences. In this research, the remaining systematic error in the RD model has been indirectly compensated by identifying the corresponding positions between the DEM and the SAR image and through simple geometric transformations in the image space [10].…”