摘要
为了解决电器芯线生产中出现的芯线包筒检测精确度不高和检测效率低的问题,提出一种基于机器视觉的非侵入式缺陷检测方法用以检测喇叭口缺失。通过图像滤波去噪、图像的二值化处理以及图像的特征提取,解决检测背景复杂、噪声干扰大等问题,并提高图像的对比度、突出感兴趣区域(ROI),再采用梯度算法进行边缘特征检测,进而对缺陷区域特征信息进行快速的定位及分类。试验结果表明,该缺陷检测方法不仅可以有效提高检测效率,又能保证较高的检测准确度和精度,满足实际工业检测需求,具有良好的实用价值。
In order to solve the problems of low accuracy and low efficiency in detecting the core wire package in the production of electrical core wires,a non-intrusive defect detection method based on machine vision is proposed to detect the absence of speaker mouths.Through image filtering,binarization,and feature extraction,the method solves problems such as complex background and large noise interference,improves the contrast of the image,highlights the region of interest(ROI),and uses gradient algorithms for edge feature detection.Then,the feature information of the defect area is quickly located and classified.Experimental results show that the defect detection method can not only effectively improve detection efficiency,but also ensure high detection accuracy and precision,which meets the needs of practical industrial detection and has good practical value.
作者
季坚莞
张胜利
陈淼
李伟
JI Jianwan;ZHANG Shengli;CHEN Miao;LI Wei(Keyi College of Zhejiang Sci-Tech University,School of Information and Control,Shaoxing Zhejiang 312300)
出处
《软件》
2023年第7期48-51,共4页
Software
基金
浙江理工大学科技与艺术学院院设科研项目资助(KY2022005)。
关键词
机器视觉
芯线包筒
缺陷检测
图像处理
非侵入式
machine vision
core wire sheathing
defect detection
image processing
non-intrusive
