Article ID Journal Published Year Pages File Type
6949072 ISPRS Journal of Photogrammetry and Remote Sensing 2018 6 Pages PDF
Abstract
Geometric calibration can eliminate systematic errors in the imaging parameters to improve the positioning accuracy. The conventional method depends on high-accuracy control data to detect and eliminate systematic errors; therefore, it requires a satellite to scan the calibration fields before carrying out geometric calibration, which reduces its timeliness. This paper presents an interior self-calibration method to eliminate CCD distortions without using high-accuracy control data on the calibration fields, in contrast to the conventional calibration method. Instead, only the 90-m digital elevation data from the global shuttle radar topography mission (90m-SRTM) are used. Extraordinarily, the calibration accuracy of this method is not restricted by the inadequate accuracy of 90m-SRTM, and very promising accuracy comparable to that of the conventional calibration method is achieved. In the experiment, multispectral images obtained by the ZY02C satellite were used, and the results show that the proposed method can eliminate the high-order distortion without the use of high-accuracy control data but only with the 90m-SRTM. Moreover, the internal accuracies of the calibrated images are better than 0.8 pixels. Compared to the conventional method using high-accuracy control data for calibration, the difference between the two methods is only about 0.02 pixels, which verifies the effectiveness of this method.
Related Topics
Physical Sciences and Engineering Computer Science Information Systems
Authors
, , , , , , ,