Adaptive threshold discriminating algorithm for remote sensing image corner detection

An adaptive threshold discriminating Algorithm for remote sensing image corner detection was discussed in this paper. Firstly, this paper proposed a novel corner detection method, which discriminated the direction of corner by analyzing eight neighborhood direction gray gradients, then adopted the neighborhood gray gradient tracking method and two thresholds of gray gradient was adopted to detect corner point. By this corner detection method, corner of remote sensing image could be extracted correctly. Secondly, an adaptive threshold discriminating method for remote sensing image corner detection was introduced in this paper, the threshold of discriminant function could be determined adaptively by calculating probability density curvature extremum of gray gradient. By comparison with traditional empirical threshold methods, this adaptive threshold discriminating method was more intelligent and automated. The experiment illustrated that, the result of corner point detection was more objective and dependable, and it had more detection accuracy and efficiency. Meanwhile, the algorithm could detect suitable corner point, and it had more adaptability and applicable prospect.