摘要
圆形目标检测在各个领域得到了广泛应用.为了提高规则圆的检出效率、降低不规则圆的漏检率,文章提出了一种改进随机Hough变换的圆形目标检测算法.算法首先用区域增长和RGB二值分割等算法对不同背景的图像进行二值分割和去噪处理,接着对连通域分割的特征区域用具有圆形性特征的边缘点进行候选圆的检测和验证.实验对比结果表明,本文算法可以显著提高规则圆的检出效率、降低不规则圆的漏检率,算法计算量小、运行速度快、鲁棒性强.
Circular target detection has been widely used in various fields.In order to improve the detection efficiency of regular circles and reduce the missed detection rate of irregular circles,a circular target detection algorithm based on improved random Hough transform is proposed.Firstly,region growing and RGB binary segmentation are used to segment and denoise images with different backgrounds.Secondly,edge points with circular features in regions of the connected domain segmentation are used to detect and verify candidate circles.The comparison results show that the algorithm could significantly improve the detection efficiency of regular circles and reduce the missed detection rate of irregular circles.The algorithm has the advantages of low computation complexity,fast running rate and strong robustness.
作者
徐炜君
XU Weijun(Department of Electrical Information Engineering,Northeast Petroleum University at Qinhuangdao,Qinhuangdao 066004,China)
出处
《杭州师范大学学报（自然科学版）》
CAS
2024年第1期92-98,共7页
Journal of Hangzhou Normal University(Natural Science Edition)
基金
黑龙江省属本科高校引导性创新基金项目(2020YDQ-01).
关键词
背景分割
特征区域
随机HOUGH变换
圆检测
background segmentation
feature region
random Hough transform
circle detection
