基于Canny-Franklin矩的亚像素边缘检测方法  

Sub-pixel edge detection method based on Canny-Franklin moments

作　　者：杜港 佟强 侯凌燕[1] 杨大利[1] 梁旭 DU Gang;TONG Qiang;HOU Lingyan;YANG Dali;LIANG Xu(School of Computer,Beijing Information Science and Technology University,Beijing 100101,China)

机构地区：[1]北京信息科技大学计算机学院,北京100101

出　　处：《计算机集成制造系统》2025年第3期903-912,共10页Computer Integrated Manufacturing Systems

基　　金：北京信息科技大学“青年骨干教师”支持计划(YBT202426)。

摘　　要：针对现有的矩方法在亚像素边缘检测中存在执行时间较长的问题,提出一种基于Canny-Franklin矩的亚像素边缘检测算法,缩短了执行时间,其时间优势主要在于只需要2个卷积模板。首先,对Franklin矩求解亚像素边缘进行了原理推导,并计算出Franklin矩的7×7模板;然后,提出基于Canny-Franklin矩的亚像素边缘检测算法,即先对图像进行Canny边缘检测粗定位,再将粗定位边缘像素与矩模板进行卷积,并利用推导出的相关公式计算出亚像素边缘值。实验结果表明,该算法可以实现对图像边缘的亚像素定位,且在定位精度上相对Zernike矩方法具有一定的优势;经过对实际定位销钉的检测,该算法的定位结果满足实际情况,结果与Canny-Zernike矩方法结果非常接近,但算法运行时间仅为Canny-Zernike矩算法的66%,传统Zernike矩算法的50%,从而证明了所提算法在图像亚像素边缘定位中具有一定的优势。Aiming at the problem that the existing moment method has a long execution time in subpixel edge detection,a subpixel edge detection algorithm based on Canny-Franklin moment was proposed that shortened the execution time,and its time advantage mainly came from the fact that only 2 convolutional templates were needed.The principle derivation for the subpixel edge of Franklin moment solution was carried out,and the 7×7 template of the Franklin moment was calculated.Then,a subpixel edge detection algorithm based on Canny-Franklin moment was proposed that the Canny edge detection coarse positioning of the image was first performed,and then the coarsely positioned edge pixels were convolved with the moment template,and the subpixel edge value was calculated by using the derived relevant formula.Experimental results showed that the proposed algorithm could achieve subpixel positioning of image edges,and had certain advantages over Zernike moment method in terms of positioning accuracy.After testing the actual dowel pin,the positioning result of the algorithm met the actual situation,and the result was very close to the result of the Canny-Zernike moment method,but the algorithm running time was only 66%of the Canny-Zernike moment algorithm and 50%of the traditional Zernike moment algorithm.Therefore,the proposed algorithm had certain advantages in image subpixel edge positioning.

关 键 词：亚像素 边缘检测 Franklin矩 视觉测量 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]