基于曲率多尺度的高精度角点检测

为了提高角点检测的精度,满足算法的稳定性与实时性,提出了一种基于曲率多尺度和局部迭代的亚像素角点检测算法。首先基于Canny算子和自定义边缘跟踪算法检测目标轮廓,然后利用多尺度曲率多项式得到像素级的角点,最后以像素级角点为中心,利用向量正交性原则,采用迭代算法及双线性插值得到子像素级角点。实际测试表明:该方法检测效率高,提取到的角点稳定,精度可以达到0.2pixel,可满足实际应用中的实时、高精度要求,目前已成功地应用于嵌入式机器视觉工业现场。

High precision corner detection based on curvature multi-scale

In order to improve the precision of corner detection,a subpixel corner detector based on curvature multi-scale and local iteration is presented to satisfy the algorithm's stability and real-time. First,an self-defined edge tracing algorithm and the Canny algorithm are used to extract the object contour.Secondly,pixel corners are calculated by using the multi-scale curvature polynomial(MCP). Finally,to center around the pixel corner,subpixel corner is checked based on terated algorithm and bilinear interpolation with principle of vector's orthogonality.The results show that the proposed method offers a high effective and robust solution to images,the precision can achieve 0.2 pixel.It can meet the real-time,high accuracy requirement in practice. At present,the detector has been used in the industrial field of embedded machine vision.