Preliminary Analysis of Gaofen-1 B/C/D Satellite Stereo Mapping Performance
- 15 Downloads
The Gaofen-1 (GF-1) B/C/D is optical satellite with two 2/8 m spatial resolution panchromatic/multispectral cameras, and they were launched in March 2018 on a CZ-4C rocket. This paper used the GF-1 B/C/D and Ziyuan-3 (ZY-3) data that had been repeatedly visited in Northeast China to carry out mapping performance research. According to the sensor parameters and imaging geometric model, the instantaneous field of view at different side angles was analyzed, and the influence of ground control data error on image adjustment accuracy was simulated. Then, using the GF-1 B/C/D and ZY-3 data acquired simultaneously, multiple stereo models were constructed. Based on high-precision ground reference dataset, a large number of ground control points were automatically extracted for each stereo model, and then the affine model in image space and the provided rational polynomial coefficients were used for the adjustment. The results show that the root-mean-square error of the elevation of the GF-1 B/C/D stereo models is about 2–3 m. Subsequently, using these stereo models, the digital surface model (DSM) was extracted and analyzed. In general, whether it is the accuracy of stereo model or the performance of DSM extracted, it could meet the 1:50,000 scale mapping specification requirements. However, it is found that some inter-chip stitching errors between CCDs cannot be ignored in the GF-1 B/C/D image data, “elevation fracture” and abnormal elevation values are obvious in the DSM, and these defects need continuous improvement.
KeywordsGaofen-1 B/C/D ZY-3 Stereoscopic Accuracy DSM
- 1.Jacobsen K (2016) Analysis and correction of systematic height model errors. Int Arch Photogramm Remote Sens Spatial Inf Sci XLI-B1:333–339Google Scholar
- 2.Jacobsen K (2017) Problems and limitations of satellite image orientation for determination of height models. Int Arch Photogramm Remote Sens Spatial Inf Sci XLII-1/W1:257–264Google Scholar
- 3.Jacobsen K (2018) Systematic geometric image errors of very high resolution optical satellites. Int Arch Photogramm Remote Sens Spatial Inf Sci XLII-1:233–238Google Scholar
- 4.Grodecki, J (2001) IKONOS stereo feature extraction—RPC approach. In: ASPRS annual conference, St. LouisGoogle Scholar