机械零件直线边缘亚像素定位方法研究  

Study on Beeline Edge Subpixel Localization for Mechanical Part

作　　者：崔彦平[1,2] 葛杏卫[1] 张洪亮[1] 

机构地区：[1]河北科技大学机械电子工程学院,石家庄050018 [2]天津大学精密测试技术及仪器国家重点实验室,天津300072

出　　处：《半导体光电》2010年第5期797-800,806,共5页Semiconductor Optoelectronics

基　　金：教育部博士点专项科研基金资助项目(20030056017);河北科技大学科学研究基金与大学生科技创新基金资助项目

摘　　要：直线是视觉图像中重要的基元特征。建立了三级灰度图像边缘模型的空间矩算子,利用LOG算子定位速度快的特点,确定图像像素级边缘,然后在包含边缘点的邻域内利用空间矩进行边缘的亚像素定位,由Hough变换提取直线边缘像素点,最后采用基于最小二乘原理的直线拟合边缘提取方法,得到亚像素级被检测直线。对空间矩算子亚像素定位算法与像面直线亚像素提取算法的有效性和精度进行了实验研究,实验结果表明提出的基于空间矩亚像素边缘定位算法与像面直线亚像素提取算法具有较高的精度和稳定性。Beeline is the important element in the computer vision measurement.Subpixel detection of beeline on image plane based on spatial moment and 3Dreconstruction of spatial beeline is researched.First the LOG operator was used to locate the edge at the level of pixel quickly,and then the spatial moment was used to detect subpixel location of the discrete points of edge.The least square fitting was adopted to fit the subpixel edge points which had been detected by Hough transform.The validity and precision of spatial moment subpixel edge detection and the extraction of beeline on image plane was studied.The experiments show that spatial moment subpixel edge detection and the extraction of beeline on image plane have acquired high precision.

关 键 词：边缘检测 直线提取 LOG算子 空间灰度矩 HOUGH变换 最小二乘算法 

分 类 号：TP242.6[自动化与计算机技术—检测技术与自动化装置]