Ground Control Point Extraction Algorithm for Remote Sensing Image Based on Adaptive Curvature Threshold
Deng Xiaolian, Yuehua Huang, Feng Shengqin, Wang Changyao · 2008
An adaptive ground control point extraction algorithm for remote sensing image was introduced in this paper. The main idea was to extract corner of remote sensing image automatically and intelligently. This paper proposed a new corner extraction algorithm, which confirmed the direction of corner by analyzing eight direction gray gradients, then adopted the neighborhood gray gradient tracking method and uses two thresholds of gray gradient to detect the correct corner. By this corner extraction method, ground control point of remote sensing image could be extracted correctly. Meanwhile, the threshold of discriminant function could be determined adaptively by calculating probability density curvature extremum of gray gradient instead of traditional experiential threshold. The result of the experiment demonstrated that, the algorithm could extract valuable ground control point, it had more extraction accuracy and efficiency, and it had more adaptability and applicable value. The result of ground control point extraction was more objective and dependable.