基于投票约束的Hough变换直线检测算法  

Hough transform line detection algorithm based on voting constraints

作　　者：张喜民 詹海生[2] 余奇颖 ZHANG Ximin;ZHAN Haisheng;YU Qiying(School of Information Engineering,Shaanxi Institute of International Trade&Commerce,Xi’an 712046,China;School of Computer Science and Technology,Xidian University,Xi’an 710071,China)

机构地区：[1]陕西国际商贸学院信息工程学院,陕西西安712046 [2]西安电子科技大学计算机学院,陕西西安710071

出　　处：《电子设计工程》2023年第19期89-93,共5页Electronic Design Engineering

基　　金：陕西省自然科学基础研究计划(2021JM539);陕西高校青年创新团队计划(SJT23TD02);物联网与智能技术科技创新团队基金(SSY18TD05)。

摘　　要：图像边缘直线检测的精度直接影响机器视觉尺寸测量的精度,Hough变换算法用于类矩形小插件边缘直线检测时,易在Hough空间形成伪峰、影响边缘检测精准度,为此,提出了基于投票约束的Hough变换直线检测算法,通过引入Hough空间投票权重分配机制来抑制伪峰的产生,对算法性能影响较小,同时较明显地提高了检测精度。基于高分辨率工业相机和高性能计算机构建了精密测量系统,并以手机尾插零件为对象进行实验,实验结果表明,该文算法直线检测精准度可达到89%,与标准Hough算法、概率Hough算法相比较,该文算法的直线检测精准度有较大的提升,且该文研究已应用于某型检测设备研制中。The accuracy of image edge line detection directly affects the accuracy of machine vision scale measurement.Hough transform algorithm is widely used in image line detection,but the algorithm is easy to form pseudo peaks in the Hough space for small rectangle⁃like parts,which affects the accuracy of edge line detection.Therefore,voting⁃constrained Hough transform line detection algorithm is proposed:the generation of spurious peaks is suppressed by introducing voting weight distribution mechanism in Hough space,which significantly improved the detection accuracy with less affecting the performance of the algorithm.A precision measurement system is developed based on high⁃resolution industrial camera and high⁃performance computer,and experiments are carried out on mobile phone tail plug parts.The experimental results show the linear detection accuracy of the proposed algorithm can reach 89%,which has a great improvement over the standard Hough algorithm and the probabilistic Hough algorithm,and the research in this paper was applied to the development of a certain type of testing equipment.

关 键 词：机器视觉 尺寸测量 直线检测 HOUGH变换 

分 类 号：TN06[电子电信—物理电子学]