PDF全文下载:
201902015.pdf
文章编号: 16726987(2019)02010206; DOI: 10.16351/j.16726987.2019.02.015
曹梦龙, 霍立斌
(青岛科技大学 自动化与电子工程学院,山东 青岛 266061)
摘要: 针对轮胎钉孔检测以手工或X光所造成的辐射强、定位不准的问题,设计以机器视觉技术辅助高压放电的检测系统,有效检测钉孔并准确定位。采用大津法(Otsu)完成灰度变换后的图像二值化,并通过跟踪特定颜色的对象跟 踪器精确定位钉孔位置。经5组实际测试可准确地识别定位轮胎表面25 mm及以上直径通孔,检出率达92%。
关键词: 轮胎钉孔检测; 机器视觉; 灰度变换; 大津法
中图分类号: TP 216文献标志码: A
引用格式:曹梦龙, 霍立斌. 轮胎钉孔视觉检测系统设计\[J\]. 青岛科技大学学报(自然科学版), 2019, 40(2): 102107.
CAO Menglong, HUO Libin. Design of visual inspection system of tire nail hole\[J\]. Journal of Qingdao University of Science and Technology(Natural Science Edition), 2019, 40(2): 102107.
Design of Visual Inspection System of Tire Nail Hole
CAO Menglong, HUO Libin
(College of Automation and Electronic Engineering,Qingdao University of Science and Technology,Qingdao 266061,China)
Abstract: For the problem that the nail hole detection of the tire is caused by manual or Xray radiation and the positioning is not accurate, a machine vision technologyassisted highvoltage discharge detection system is designed to effectively detect the nail hole and accurately locate the nail hole.The image is binarized after the gradation transformation is performed by the Otsu , and the position of the nail hole is accurately positioned by tracking a specific color object tracker.After 5 groups of actual tests can accurately identify the location of the tire surface diameter of 2.5 mm and above through holes, the detection rate of 92%.
Key words:tire hole detection; machine vision; grayscale transformation; otsu method
收稿日期: 20180630
基金项目: 国家自然科学基金面上项目(61773227);江苏省科技计划项目(BC2015046).
作者简介: 曹梦龙(1971—),男,副教授.*通信联系人.