基于机器视觉的PET瓶坯缺陷检测系统设计  

Design of PET Bottle Preform Defect Detection System Based on Machine Vision

作　　者：吴宇翔 郑兆启 石朴 侍炳鉴 张超 李占勇[1,2] 王瑞芳[1,2] 徐庆[1,2] WU Yuxiang;ZHENG Zhaoqi;SHI Pu;SHI Bingjian;ZHANG Chao;LI Zhanyong;WANG Ruifang;XU Qing(College of Mechanical Engineering,Tianjin University of Science and Technology,Tianjin 300222,China;Tianjin Key Laboratory of Integrated Design and On-line Monitoring for Light Industry&Food Machinery and Equipment,Tianjin University of Science and Technology,Tianjin 300222,China)

机构地区：[1]天津科技大学机械工程学院,天津300222 [2]天津科技大学天津市轻工与食品工程机械装备集成设计与在线监控重点实验室,天津300222

出　　处：《机械制造与自动化》2024年第5期255-258,共4页Machine Building & Automation

基　　金：广东省重点领域研发计划项目(2020B0202010004)。

摘　　要：针对PET瓶坯缺陷人工检测工作强度大、效率低等问题,基于机器视觉技术设计一套由检测组件、控制组件、输送机构组件以及剔除机构组件等组成的PET瓶坯检测系统。通过Halcon软件对采集的PET瓶坯图像采用中值滤波降噪、自适应阈值方法提取检测区域,并使用改进Canny算法进行缺陷检测。试验结果表明:PET瓶坯检测系统可完成对PET瓶表面黑点、划痕类缺陷检测,试验检测准确率可达97.4%,可以稳定识别不合格瓶坯。Based on machine vision technology,a PET bottle blank detection system composed of detection component,control component,conveying mechanism component and eliminating mechanism component was designed to improve the high intensity and low efficiency of manual detection of PET bottle blank defects.Halcon software was used to extract the detection area of the collected PET billet images by means of median filter denoising and adaptive threshold,and the improved Canny algorithm was applied for defect detection.The test results show that the PET bottle blank detection system can complete the detection of PET bottle surface defects such as black spots and scratches,with the test detection accuracy as high as 97.4%and the cabability of stably identifying unqualified bottles.

关 键 词：瓶坯 机器视觉 阈值分割 改进Canny算子 缺陷检测 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]