Public DEM-assisted Positioning Method for Chinese Satellite Imagery without Ground Control Points

• Main Authors: CHEN Xiaowei, ZHANG Baoming, ZHANG Tonggang, GUO Haitao, CEN Minyi
• Format: Article
• Language: zho
• Published: Surveying and Mapping Press 2016-11-01
• Series: Acta Geodaetica et Cartographica Sinica
• Subjects: global mapping; positioning without ground control point; positioning assisted by DEM; DEM matching; least trimmed squares estimator; distribution function of height difference
• Online Access: http://html.rhhz.net/CHXB/html/2016-11-1361.htm

With the purpose of revolving the problem of photogrammetry without ground control points (GCPs) (referred to non-controlled positioning), a public DEM-assisted positioning method for Chinese satellite imagery without GCPs was proposed in the context of global mapping. To give full play to the characteristics of globally consistent, stable and high precision of public DEM, these data were regarded as reference data and matched with the DEM extracted from satellite imagery. Least Z-difference (LZD) method using least trimmed square (LTS) estimator was applied to match, and deformation between DEM is automatically detected and eliminated by analysis of the distribution of Z-difference. The correction parameters in object space which determined by DEM matching were applied to correct the direct positioning results. Multi-groups of comparative experiments for Mapping Satellite-1 and ZY-3 Satellite imagery were designed. Experimental results show that the proposed method is practical and feasible and has good robustness to gross error and deformation between surfaces, which can make full use of the advantages of the reference DEM. The positioning accuracy greatly depends on but not limited to the accuracy of public DEM and hardly affected by its resolution. Assisted by SRTM DEM, single imagery can realize high positioning accuracy without GCPs, which can meet the mapping standard of 1:50,000 scale. This method also provides a new and effective means for positioning accuracy inspection of satellite imagery.