Block Adjustment without GCPs for Chinese Spaceborne SAR GF-3 Imagery

Area of Science:

• Remote Sensing
• Geomatics Engineering
• Photogrammetry

Background:

• Gaofen-3 (GF-3) satellite provides C-band multi-polarization synthetic aperture radar (SAR) data crucial for high-accuracy mapping in China.
• Current GF-3 SAR imagery processing relies heavily on Ground Control Points (GCPs) for accurate mapping, posing a significant limitation.
• Error sources affecting geometric processing of GF-3 SAR imagery require thorough analysis to improve mapping precision.

Purpose of the Study:

• To develop and validate a novel block adjustment method for GF-3 SAR imagery that does not require Ground Control Points (GCPs).
• To analyze and mitigate error sources impacting the geometric processing of GF-3 SAR data.
• To enhance the geometric accuracy and absolute positioning of GF-3 orthophotos for broader applications.

Main Methods:

• Geometric calibration of GF-3 SAR images was performed.
• A Rational Polynomial Coefficient (RPC) model was generated post-calibration, incorporating compensated geometric parameters.
• A Digital Elevation Model (DEM) assisted planar block adjustment was employed to solve orientation parameters and conduct ortho-rectification.

Main Results:

• The proposed block adjustment method successfully processed GF-3 SAR imagery from QPSI and FS2 imaging modes without GCPs.
• Experimental results demonstrated geometric mosaic accuracy better than one pixel.
• Absolute positioning accuracy of the resulting orthophotos also exceeded one pixel.

Conclusions:

• The developed GCP-free block adjustment method significantly improves the geometric accuracy of GF-3 SAR imagery.
• The achieved sub-pixel accuracy provides a robust foundation for GF-3 data utilization in global high-accuracy mapping initiatives.
• This method reduces reliance on ground control, streamlining SAR data processing workflows.