摘要
针对传统人工检测铣刀崩刃和多刃缺陷时效率低下、精度不高等问题,提出一种基于形态学多曲率融合的铣刀崩刃、多刃缺陷检测方法。首先,使用色彩阈值分离前后景;然后,使用双边滤波对图像进行滤波处理;随后,二值化图像并去除小连通域的干扰,通过边界跟踪得到刀刃轮廓;最后,使用改进的链码计算机制进行刀刃轮廓曲度的计算,并通过曲度定位缺陷点。经过实验验证,该检测方法具有平均0.8 s每样本的检测速度和92%的检测精度,能满足工业生产在线检测的要求。
In response to the problems of low efficiency and low accuracy in the traditional manual detection of milling cutter breakage and multi edge defects,a milling cutter breakage and multi edge defect detection method based on morphological multi curvature fusion is proposed.First,color threshold is used to separate positive and negative scenes.Second,the bilateral filter is used to filter the image.Then,the image was binarized to remove the interference of small connection domain,and obtain the blade contour through boundary tracking.Finally,the curvature of blade profile is calculated by an improved chain coding computer system,and defect points are located through the curvature.After experimental verification,this detection method has an average detection speed of 0.8 s per sample and a detection accuracy of 92%,which can meet the requirements of industrial production online detection.
作者
易忠
周骅
赵麒
袁学枫
YI Zhong;ZHOU Hua;ZHAO Qi;YUAN Xuefeng(College of Big Data and Information Engineering,Guizhou University,Guiyang 550025,China;College of Mechanical and Electronic Engineering,Guizhou Minzu University,Guiyang 550025,China)
出处
《组合机床与自动化加工技术》
北大核心
2024年第2期155-159,共5页
Modular Machine Tool & Automatic Manufacturing Technique
基金
国家自然科学基金联合基金重点支持项目(U1836205)
贵州大学培育项目(黔科合平台人才[2017]5788-60)。
关键词
铣刀
缺陷检测
高精度检测
计算机视觉
图像处理
milling cutter
defect detection
high-precision detection
computer vision
image processing