摘要
油画棒装盒传统方式采用人工,针对其摆放效率低、成本高的问题,设计了基于机器视觉的油画棒装盒系统,选用ARM芯片做图像处理,在有限的资源配置和时间条件下完成任务。系统通过摄像头对油画棒进行图像采集,控制电机转动油画棒,进行图像匹配,完成投放。系统先对图像定位、选取,滤波去噪,再采用改进的自适应阈值的直方图匹配算法,并结合模板图像和待匹配图像的相关系数,有效降低了计算量,提高了系统的识别率。实践证明系统的工作稳定,能实现对油画棒角度的有效调整,匹配速度快,单台识别率达到99%以上,能完成产品的自动装配。
The traditional way of oil painting stick packing is artificial.In order to solve the problem of low efficiency and high cost,this paper designs the system based on machine vision.The ARM chip is chosen as coreof the image processing system,under the condition of limited resource and time.The system collects the image through the camera,controls the motor to rotate the oil painting stick,carries on the image matching,completes delivers.The system first locates,selectsthe image and filters to denoise the image.Thenthe system uses the improved histogram matching algorithm of adaptive threshold,and combines the correlation coefficient between the template image and the image to be matched,which effectively reduces the computation and improves the recognition rate of the system.Practice shows that the work of the system is stable,can achieve the effective adjustment of the angle of the oil paintingwith a high matching speed.The recognition rate of thesystem is above 95%.The system can complete the automatic assembly of products.
出处
《电子测量技术》
2017年第7期121-125,共5页
Electronic Measurement Technology
基金
上海市自然科学基金(15ZR1417400)
国家自然科学基金青年基金(61302181)资助项目
关键词
机器视觉
图像处理
直方图匹配
相关系数
machine vision
image processing
histogram matching
correlation coefficient
