[1]于晓海,张 阳,须 颖.一种改进自适应阈值的 Canny算法[J].机械与电子,2020,(01):6-9.
 ,An Improved Adaptive Threshold Canny Algorithm[J].Machinery & Electronics,2020,(01):6-9.
点击复制

一种改进自适应阈值的 Canny算法()
分享到:

机械与电子[ISSN:1001-2257/CN:52-1052/TH]

卷:
期数:
2020年01期
页码:
6-9
栏目:
设计与研究
出版日期:
2020-01-24

文章信息/Info

Title:
An Improved Adaptive Threshold Canny Algorithm
文章编号:
1001- 2257(2020)01- 0006- 04
作者:
于晓海张 阳须 颖
沈阳建筑大学机械工程学院,辽宁 沈阳 110168
Author(s):
YUXiaohaiZHANGYangXUYing
SchoolofMechanicalEngineering,ShenyangJianzhuUniversity,Shenyang110168,China
关键词:
Canny算子边缘检测双边滤波Otsu法最优化阈值分割法信噪比
Keywords:
CannyedgedetectoredgedetectionbilateralfilterOstuoptimalthresholdsegmentationSNR
分类号:
TP391.41
文献标志码:
A
摘要:
:针对传统 Canny算法在滤波时无差别平滑降噪、边缘信息丢失及模糊的现象,且人工选取梯度高低阈值存在局限性的问题,提出了综合考虑图像位置信息与亮度信息的双边滤波代替传统高斯滤波方法,结合最优化阈值分割法和 Otsu法分别计算出图像高低阈值,采用边缘信噪比SNR函数对比验证边缘检测效果。改进后的 Canny算法有效提高了边缘完整性,且伪边缘更少,对图像中相对较弱的边缘部分也有
良好的提取效果,改进后的算法具有更高的信噪比。采用工业机械臂视觉采集零件边缘为背景,实验结果表明,改进后的 Canny算法可以有效保留真实边缘信息,边缘更加连续且完整。
Abstract:
Aiming at the problem that the traditional Canny algorithm has no difference in smoothing and other problems such as noise reduction edge information loss and blurring during filtering, and the limitation of artificially selecting the gradient high and low threshold, a bilateral filtering that comprehensively considers image position information and luminance information was proposed instead of the traditional Gaussian filtering method. The optimal threshold segmentation method and the Otsu method were used to calculate the image height threshold, and the edge signal-to-noise ratio SNR function was used to verify the edge detection effect. The improved Canny algorithm effectively improves edge integrity, and has fewer false edges, and also has a good extraction effect on relatively weak edge portions in the image. The improved algorithm has a higher signal-to-noise ratio; using industrial machinery. The arm visually captures the edge of the part as the background. It is proved by experiments that the improved Canny algorithm can effectively preserve the true edge information, and the edge is more continuous and complete.

参考文献/References:

[1] MUTHUKRISHNANR,RADHA M.Edgedetection tetectiontechniquesforimagesegmentation[J].InternationalJournalofComputerScienceandInformation Technology,2012,3(6):250- 254.

[2] SHRIVAKSHAN G T.Comparisonofvariousedge detectiontechniquesusedinimageprocessing[J].InternationalJournalofComputerScienceIssues,2012,9(5):269- 276.
[3] MARR D,HILDRETH E.Theoryofedgedetection[J].ProceedingsoftheRoyalSocietySeriesB:BiologicalSciences,1980,207(1167):187- 217.
[4] CANNYJ.Acomputationalapproachtoedgedetection[J].IEEE TransactionsonPattern Analysisand MachineIntelligence,1986,8(6):679- 698.
[5] 侯 守 明,王 阳,唐 琪 博,等.一 种 基 于 偏 微 分 方 程 和Canny算子的图像分割方法[J].山西大学学报(自然
科学版),2017,40(4):676- 682.
[6] 孙砚飞,常晓刚,李东兴,等.基于自适应 Canny的红外
图像边缘检测算法[J].山东理工大学学报(自然科学版),2017,31(6):18- 21.
[7] 刘庆民,张蕾,李雪.基于改进 Canny的芯片图像边缘
检测算法[J].计算机工程与设计,2016,37(11):3063- 3067,3117.
[8] 王能,赵红立,琚生根,等.一种改进的 Canny边缘检测
自适应算法[J].四川大学学报(自然科学版),2014,51
(3):479- 482.
[9] DEMIGNYD,LORCAFG,KESSALL.Evaluation ofedgedetectorsperformanceswithadiscreteexpressionofCanny’scriteria [C]//InternationalConferenceonImageProcessing.LosAlamitos,CA ,USA:IEEE,ComputerSOCPress,1995:169- 172.
[10] DEMIGNYD,KAMLET.Adiscreteexpressionof Canny’scriteriaforstepedgedetectprperformances evaluation[J].IEEE TransactionsonPatternAnalysisiand MachineIntelligence,1997,19(11):1199- 1211.
[11] TOMASIC,MANDUCHIR.Bilateralfilteringfor grayandcolorimages[C]//InternationalConference
onComputerVision,2002:839.
[12] OTSU N.Athresholdselectionmethodfromgraylevelhistograms[J].IEEETransactionsonSystems,ManandCybernetics,2007,9(1):62- 66.

相似文献/References:

[1]张小洁,段峻.改进的Hough变换在零件圆心定位中的应用[J].机械与电子,2016,(10):62.
 ZHANG Xiaojie,DUAN Jun.Application of Improved Hough Transforming in Parts Center Positioning[J].Machinery & Electronics,2016,(01):62.
[2]魏 哲1,张 阳1,张明明2,等.视觉引导划片机轨迹优化算法研究[J].机械与电子,2019,(12):38.
 WEI Zhe,ZHANG Yang,ZHANG Mingming,et al.Research on Trajectory Optimization Algorithm of Visual Guide Grinding Wheel Dicing Machine[J].Machinery & Electronics,2019,(01):38.
[3]王建新1,朱 煜1,2,等.船体零件几何尺寸测量图像分割方法研究[J].机械与电子,2021,(05):68.
 WANG Jianxin,ZHU Yu,HU Xiaofeng,et al.Research on Image Segmentation for Geometric Dimension Measurement of Hull Parts[J].Machinery & Electronics,2021,(01):68.
[4]高英姿,姚爱琴,孙运强,等.基于二进小波和抗噪形态学的烟雾边缘检测算法[J].机械与电子,2021,(08):18.
 GAO Yingzi,YAO Aiqin,SUN Yunqiang,et al.Smoke Edge Detection Algorithm Based on Dyadic Wavelet and Anti Noise Morphology[J].Machinery & Electronics,2021,(01):18.
[5]董博航,陈 平.基于机器视觉的硅钢板边浪平直度测量方法[J].机械与电子,2022,(12):53.
 DONG Bohang,CHEN Ping.Measurement Method of Edge Wave Straightness of Silicon Steel Plate Based on Machine Vision[J].Machinery & Electronics,2022,(01):53.

备注/Memo

备注/Memo:
收稿日期:2019- 09- 12
基金项目:辽宁省教育厅科研专项(LJZ2016018)
作者简介:于晓海 (1995-),男,内蒙古通辽人,硕士研究生,研究方向为机器人视觉识别;张 阳 (1981-),男,辽宁沈阳人,副教授,博士,研究方向为先进智能机电系统、机器人视觉伺服与移动机器人。
更新日期/Last Update: 2020-01-13