基于自适应阈值的多尺度边缘检测方法

Multi-scale Corner Detection Methodwith Adaptive Threshold

图像边缘信息在物体识别方法中具有重要作用,采用多尺度特征检测能同时检测出细微和粗糙特征。基于曲率尺度空间(CSS)技术,文章提出了一种改进的多尺度边缘检测方法,该方法首次利用自适应局部曲率阈值代替了原有CSS方法中的单一全局阈值,另外,为了消除虚假边缘点,候选边缘点的角度被检测在一组动态范围内。实验结果表明,该方法能有效解决多尺度特征的图像检测问题。

Corners play animportant role in object identificationmethods.Multi-scalefeature detection can detect both fine and coarse features at the same time.A new and improved multi-sc ale corner detectionmethodbased on Curvature Scale Space （CSS） technique is proposed in this paper.It first use an adaptive local curvature threshold instead of a single global threshold as in the original and enhanced CSS methods.For eliminating falsely detected corners,the angles of corner candidates are checked in a dy-namic region of support.The experimental results showed that the proposed method offers a robust and effective solution to images containing widely different size features.