基于机器视觉的物料自动分拣系统研究
|
摘要
为解决工业自动化生产中的自动分拣问题,以工业机器人与海康威视MV-CA013-20GM型工业智能相机为基础,设计了一个基于机器视觉的物料自动分拣系统。利用OpenCV完成了相机的标定,并基于物料位置坐标提出了图像去重复算法。对原始图像进行预处理后,以匹配子图和模板图像灰度值的相关函数、归一化相关匹配准则为基础,提出了多阈值模板匹配算法实现对分拣对象的快速识别。实验结果表明:该分拣系统稳定可行,能满足实际生产中对传送带上的物料进行自动分拣的需要,提高了生产线的自动化和智能化水平。
Aiming at the problem of automatic sorting in industrial automation production,make industrial robot and Hikvision MV-CA013-20GM series industrial intelligent camera as the foundation,to design a automatic sorting system of material based on machine vision.Camera calibration is accomplished by OpenCV,and an image duplication algorithm based on material location coordinates is proposed.A multi-threshold template matching algorithm is proposed based on the correlation function of the gray value of the matching subgraph and template image and the normalized correlation matching criterion after preprocessing the original image to realize the fast recognition of sorted objects.The results show that the sorting system is stable and feasible, which can meet the needs of automatic sorting of materials on conveyor belt in actual production and improves the level of automation and intelligence of production line.
Aiming at the problem of automatic sorting in industrial automation production,make industrial robot and Hikvision MV-CA013-20GM series industrial intelligent camera as the foundation,to design a automatic sorting system of material based on machine vision.Camera calibration is accomplished by OpenCV,and an image duplication algorithm based on material location coordinates is proposed.A multi-threshold template matching algorithm is proposed based on the correlation function of the gray value of the matching subgraph and template image and the normalized correlation matching criterion after preprocessing the original image to realize the fast recognition of sorted objects.The results show that the sorting system is stable and feasible, which can meet the needs of automatic sorting of materials on conveyor belt in actual production and improves the level of automation and intelligence of production line.
引文
[1] 王诗宇,林浒,孙一兰,等.基于机器视觉的机器人分拣系统的设计与实现[J].组合机床与自动化加工技术,2017(3):125-129.
[2] 卢振利,谢亚飞,周立志,等.基于机器视觉的机器人辨识及分拣盒装香烟的系统[J].高技术通讯,2016,26(6):585-592.
[3] 周军,蔡建,郭俊先,等.基于机器视觉的核桃仁特征提取与分级方法研究[J].江苏农业科学,2018,46(11):175-179.
[4] Tuong Phuoc Tho,Nguyen Truong Thinh.Design and Development of the Sorting System Based on Robot[C].2015 15th International Conference on Control,Automation and Systems(ICCAS),Busan,Korea:IEEE,2015:1639-1644.
[5] 徐德,谭民,李原.机器人视觉测量与控制[M].3版.北京:国防工业出版社,2016.
[6] 毛星云,冷雪飞,王碧辉,等.OpenCV3编程入门[M].北京:电子工业出版社,2015.
[7] 倪鹤鹏,刘亚男,张承瑞,等.基于机器视觉的Delta机器人分拣系统算法[J].机器人,2016,38(1):49-55.
[8] 夏群峰,彭勇刚.基于视觉的机器人抓取系统应用研究综述[J].机电工程,2014,31(6):697-701.
[9] 张开生,韦逸野.数字化功能信息纸张的产品建模与图像识别研究[J].中国造纸学报,2018,33(1):61-66.
[10] Li Y,Wang R,Cui Z,et al.Spatial Pyramid Covariance based Compact Video Code for Robust Face Retrieval in TV-series[J].IEEE Transactionson Image Processing,2016,25(12):5905-5919.
[11] 张雷.图像匹配及融合算法的研究[D].成都:电子科技大学,2016.
[12] 管经纬.基于机器视觉的工件自动分拣系统研究[D].上海:东华大学,2015.
[13] 党宏社,张超,庞毅.基于视觉引导的工业机器人快速分拣系统研究[J].电子器件,2017,40(2):481-485.
[2] 卢振利,谢亚飞,周立志,等.基于机器视觉的机器人辨识及分拣盒装香烟的系统[J].高技术通讯,2016,26(6):585-592.
[3] 周军,蔡建,郭俊先,等.基于机器视觉的核桃仁特征提取与分级方法研究[J].江苏农业科学,2018,46(11):175-179.
[4] Tuong Phuoc Tho,Nguyen Truong Thinh.Design and Development of the Sorting System Based on Robot[C].2015 15th International Conference on Control,Automation and Systems(ICCAS),Busan,Korea:IEEE,2015:1639-1644.
[5] 徐德,谭民,李原.机器人视觉测量与控制[M].3版.北京:国防工业出版社,2016.
[6] 毛星云,冷雪飞,王碧辉,等.OpenCV3编程入门[M].北京:电子工业出版社,2015.
[7] 倪鹤鹏,刘亚男,张承瑞,等.基于机器视觉的Delta机器人分拣系统算法[J].机器人,2016,38(1):49-55.
[8] 夏群峰,彭勇刚.基于视觉的机器人抓取系统应用研究综述[J].机电工程,2014,31(6):697-701.
[9] 张开生,韦逸野.数字化功能信息纸张的产品建模与图像识别研究[J].中国造纸学报,2018,33(1):61-66.
[10] Li Y,Wang R,Cui Z,et al.Spatial Pyramid Covariance based Compact Video Code for Robust Face Retrieval in TV-series[J].IEEE Transactionson Image Processing,2016,25(12):5905-5919.
[11] 张雷.图像匹配及融合算法的研究[D].成都:电子科技大学,2016.
[12] 管经纬.基于机器视觉的工件自动分拣系统研究[D].上海:东华大学,2015.
[13] 党宏社,张超,庞毅.基于视觉引导的工业机器人快速分拣系统研究[J].电子器件,2017,40(2):481-485.