首页 | 本学科首页   官方微博 | 高级检索  
     

基于树莓派的塑料纽扣次品检测系统设计
引用本文:韩雪,朱凌云.基于树莓派的塑料纽扣次品检测系统设计[J].自动化仪表,2020(1):42-45.
作者姓名:韩雪  朱凌云
作者单位:东华大学数字化纺织服装技术教育部工程研究中心;东华大学信息科学与技术学院
摘    要:针对传统纽扣产业中人工检测过程效率低、经济成本高等问题,设计了一套基于树莓派的塑料纽扣次品检测系统。该系统结合了树莓派开发板、传感器、工业相机、气动剔除装置等硬件设备与软件检测算法,形成一套完整的检测系统。通过工业电荷耦合器件(CCD)相机与传感器配合实时采集流水线上纽扣的图像信息,并将其传输到树莓派,经过图像滤波去噪后,使用局部自适应阈值法保留污渍等纽扣表面易缺失细节。通过检索纽扣轮廓信息以剔除少孔、存在污渍块的次品。通过评价轮廓圆度剔除破损纽扣,最后依据评价结果由树莓派发出控制信号剔除次品纽扣。经试验验证,该检测系统能够较好地实现纽扣次品在线检测的准确性与高效性,且由于其硬件设备体积较小、成本较低、便于维护与管理,易于安装与推广至生产线环境。

关 键 词:机器视觉  次品检测  树莓派  局部自适应阈值法  轮廓检测  圆度评价

Disign of Plastic Button Defective Product Detection System Based on Raspberry Pi
HAN Xue,ZHU Lingyun.Disign of Plastic Button Defective Product Detection System Based on Raspberry Pi[J].Process Automation Instrumentation,2020(1):42-45.
Authors:HAN Xue  ZHU Lingyun
Affiliation:(Engineering Research Center of Digitized Textile&Fashion Technology of the Ministry of Education,Donghua University,Shanghai 201620,China;School of Information Science and Technology,Donghua University,Shanghai 201620,China)
Abstract:Aiming at the problems of low efficiency and high economic cost of manual detection in traditional button industry,a set of plastic button defective product detection system based on Raspberry Pi was designed.The system was composed of hardware devices such as Raspberry Pi development board,sensors,industrial cameras,pneumatic rejection devices and software detection algorithm.Real-time image information of buttons on the production line was collected by the industrial charge coupled device(CCD) camera and the sensor,and transmitted to the Raspberry Pi.After image filtering and denoising,details of button surface defects such as stains were retained by using the local adaptive threshold method.Buttons missing holes or with stains were identified by retrieving the button contour information and broken buttons were identified by evaluating the roundness of the contour.Finally,the defective buttons were culled by the control signal sent by Raspberry Pi according to the evaluation results.It was verified by experiments that the detection system can realize on-line detection of defective buttons accurately and efficiently,and it has the advantages of small size,low cost,and convenient installation,maintenance and management.
Keywords:Machine vision  Defective product detection  Raspberry Pi  Local adaptive threshold method  Contour detection  Roundness evaluation
本文献已被 CNKI 维普 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号