Correction of the SCANSAR Scalloping

• Main Author: Masanobu Shimada
• Other Authors: The Pennsylvania State University CiteSeerX Archives
• Format: Text
• Language: English
• Subjects: Sea ice
• Online Access: http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.1061.6071; http://ap-s.ei.tuat.ac.jp/isapx/2007/pdf/3D2-3.pdf

Abstract SCANSAR is an extra mode of the SAR to widen the swath of STRIP mode in up to several times. PALSAR SCANSAR showed the great performance at the radiometry and geometry and can be fully usable for monitoring the environment, sea ice, wetland, forest, and so on. One of the artifacts that the SCANSAR is sometimes affected is the scalloping, which often appears as duplicated azimuth stripes when the azimuth antenna pattern is not determined accurately. The second artifact is the azimuth ambiguity, which appears over the hetero uniform area when the sampling theorem is not satisfied at the data acquisition. This paper describes a method to determine the azimuth antenna pattern from the SAR data using the Amazon so that the scalloping does not appear, and it also shows a method to select the filter to suppress the azimuth truncation. Some PALSAR sample data corrected shows the proof of the methodology. I. INTRODUCTION While the SCANSAR with the wider swath and thus the shorter revisit time is more effective for the earth surface observation, the image quality is degraded by the three typical noises in azimuth and range directions. They are a periodic noise in the azimuth direction, which is so called the azimuth scalloping; the truncation noise in the azimuth direction, and the banding between the two scans, which appears in the range direction. SCANSAR scalloping is an artifact that is caused by the mismatching of the real azimuth power pattern and the optimum weighting function in the azimuth direction. Unless the noise floor level or the saturation at the SAR data is high enough, the correction of the scalloping is not difficult. The correction methods of the scalloping were proposed by several authors (Bamler???). They were to build the weighting function using the real azimuth antenna pattern and the multi looking intervals. They worked properly. In this paper, we proposed a method to estimate the azimuth weighting function using the Amazon forest, since the SAR data over the Amazon are uniform enough to ...