测绘通报 ›› 2017, Vol. 0 ›› Issue (3): 22-25.doi: 10.13474/j.cnki.11-2246.2017.0076

Previous Articles     Next Articles

Accuracy Evaluation and Improvement of Volunteer Geographic Information Based on Image Matching

MA Chao, SUN Qun, XU Qing, WANG Zhijian   

  1. Institute of Geospatial Information, Information Engineering University, Zhengzhou 450001, China
  • Received:2016-05-11 Revised:2017-02-07 Online:2017-03-25 Published:2017-03-31

Abstract: Aiming at the question that it is difficult to evaluate the accuracy of the volunteer geographic information under no vector reference data condition, a road accuracy evaluation method based on imaging matching is proposed. Firstly, the road intersections between the image and the volunteer geographic information are extracted as control points, and the corresponding control points are matched,then, the root mean square error of the control points of the same name are taken as road accuracy of the volunteer geographic information; Secondly, Delaunay triangulation network are respectively built by two sets of the control points, and affine translation for each triangulation network are conducted by using the corresponding relationship between these two sets, in order to achieve geometric correction of volunteer geographic information and improve its accuracy. Finally, the road data of Open Street Map in Zhengzhou city are utilized to do an experiment. The result shows the algorithm proposed in this paper is able to improve the road accuracy of the volunteer information effectively.

Key words: image matching, volunteer geographic information, accuracy evaluation, geometric correction

CLC Number: