锻压技术

冲压磨削平板类零件的表面缺陷检测

英文标题：Surface defect detection on stamping and grinding flat parts

作者：陈麒麟王东兴林建钢田洪志

单位：烟台大学

分类号：TH166

出版年,卷(期):页码：2020,45(6):168-174

摘要：

为实现机器视觉在冲压磨削平板类零件表面缺陷检测的应用，开发了一套在线检测系统。磨削加工后，零件表面有走向一致、深浅不一的磨削刀纹。实验发现，平面光照下零件整体区域清晰、轮廓明显，但细微缺陷难以发现；有向光可以突出细微缺陷，但刀纹反光，造成图像质量较差。针对以上情况，设计了一种平面光与有向光结合的打光方式，以利用他们各自的优点。根据缺陷分布特征将零件区域分为内、外两部分分别寻找缺陷，采用模板匹配、阈值分割、特征提取等方式进行图像处理。针对刀纹反光影响，提出了一种基于边缘寻找缺陷的方法，根据灰度异常区域与边缘线的位置、方向判断该异常是由缺陷引起还是由反光引起。经实验验证，本方法检测的准确率可达97%。

In order to realize the application of machine vision in surface defect detection of stamping and grinding flat parts, the on-line inspection system was developed. After the grinding process, the grinding blade patterns on the surface of parts had the same direction and different depth.It was found that the clear whole area and the obvious outline of part under plane illumination were observed by experiment, but the subtle defects were difficult to find. Then, the directed light highlighted the subtle defects, but the blade pattern was reflective resulting in poor image quality. For the above situations, the lighting method combining plane light and directed light was designed to take advantage of their respective advantages. According to the characteristics of defect distribution, the part area was divided into internal and external parts to find the defects, and the image was processed by template matching, threshold segmentation and feature extraction. Furthermore, for the influence of blade pattern reflection, the method of finding defects based on edges was proposed, and it was determined whether the abnormality was caused by defect or by light reflection according to the positions and directions for the abnormal grayscale regions and the edges. The experiments show that the accuracy of this method is up to 97%.

基金项目：

国家自然科学基金青年项目（11604285）

陈麒麟（1994-），男，硕士研究生 E-mail：501673549@qq.com 通讯作者：王东兴（1964-），男，博士，教授 E-mail：dxwang@ytu.edu.cn

参考文献：

［1］李松，周亚同，张忠伟，等.基于双打光模板匹配的冲压件表面缺陷检测
［J］.锻压技术，2018，43（11）:137-145.

Li S，Zhou Y T，Zhang Z W，et al. Surface defect detection of stamping parts based on double light pattern matching
［J］.Forging & Stamping Technology,2018,43(11):137-145.

［2］陈广峰，管观洋，魏鑫.基于机器视觉的冲压件表面缺陷在线检测研究
［J］.激光与光电子学进展，2018，55（1）：335-341.

Chen G F,Guan G Y,Wei X. Online stamping parts surface defects detection based on machine vision
［J］.Laser & Optoelectronics Progress,2018,55(1): 335-341.

［3］李克斌，余厚云，周申江.基于形态学特征的机械零件表面划痕检测
［J］.光学学报，2018，38（8）:260-266．

Li K B,Yu H Y,Zhou S J.Surface scratch detection of mechanical parts based on morphological features
［J］. Acta Optica Sinica, 2018，38（8）:260-266.

［4］徐信. 冲压件表面缺陷图像检测系统的研究与开发
［D］. 广州：广东工业大学，2016.

Xu X.Stduy and Development of Surface Defect Detecting System of Stamping Parts Based on Image Processing
［D］. Guangzhou:Guangdong University of Technology, 2016.

［5］孙义帅, 孙红星,张超,等.汽车转向轴齿磨削裂纹产生的机理及预防措施
［J］.锻压技术,2019,44(7):92-96.

Sun Y S,Sun H X,Zhang C, et al.Mechanism and preventive measures on grinding cracks for automobile steering shaft tooth The mechanism and preventive measures of grinding carck of automobile steering shaft teeth
［J］.Forging & Stamping Technology,2019,44(7):92-96.

［6］李永敬，朱萍玉，孙孝鹏，等.基于形状模板匹配的冲压件外形缺陷检测算法研究
［J］.广州大学学报：自然科学版，2017，16（5）：62-66.

Li Y J,Zhu P Y,Sun X P,et al.Shape defect detection algorithm of stamping parts based on shape template matching
［J］. Journal of Guangzhou University：Natural Science Edition,2017,16(5):62-66.

［7］Chaumette F. Image moments: A general and useful set of features for visual servoing
［J］. IEEE Transactions on Robotics and Automation, 2004, 20(4):713-723.

［8］刘馨，周彬，秦玉娟.基于Halcon的金属零件矩形度检测算法研究
［J］.自动化与仪器仪表，2017，（7）：31-32，35.

Liu X,Zhou B,Qing Y J.Halcon’s metal part rectangledetection algorithm research
［J］. Automation & Instrumentation,2017，(7): 31-32，35.

［9］李永婷，夏琴香，肖刚锋，等.基于机器视觉的锥形旋压件起皱缺陷在线检测方法
［J］.锻压技术,2019,44(1):134-141.

Li Y T,Xia Q X,Xiao G F,et al.On-line detection me-thod for wrinkling defect of conical spinning parts based on machine vision
［J］.Forging & Stamping Technology,2019,44(1):134-141.

［10］陈英. 机器视觉检测技术在工业检测中的应用
［J］．电子测试，2015，(18):79-80.

Chen Y． Machine vision inspection technology in industrial application detection
［J］.Electronic，2015，(18):79-80.

［11］吴雅莎,肖小亭,陈康,等.基于机器视觉的弯曲角度检测
［J］.锻压技术,2018,43(8):184-189.

Wu Y S,Xiao X T,Chen K,et al.Bending angle detection based on machine vision
［J］.Forging & Stamping Technology,2018,43(8):184-189.

［12］刘洋,陆倚鹏,高嵩,等.边缘检测在盘形悬式瓷绝缘子串红外图像上的应用
［J］.电瓷避雷器,2020,(1):198-203.

Liu Y ,Lu Y P,Gao S,et al.The application of edge detection in the infrared image of disc suspension insulator string
［J］.Electromagnetic Arrester,2020,(1):198-203.

服务与反馈：

【文章下载】【加入收藏】