Tie-point Automatic Matching of Multi-Sensor Images using Edge Detection base on Thinning Algorithm

摘要

This paper introduces an automatic tie-point matching algorithm which can be applied to multi-sensor satellite and air-borne image pairs such as IKONOS, SPOT, and frame camera images. This method is based on several image processing techniques such as pre-processing, filtering, edge thinning, interest point extraction, cross-correlation, and so on, in order to optimize the matching accuracy and speed by taking full characteristics of multi-sensor images (different geometry, different radiometric characteristics, etc) into account. In addition, the proposed algorithm is combined with a sensor model for minimizing search area and eliminating false-matching points automatically.