摘要
提出了一种基于单目视觉的传送带上产品定位技术.该技术利用相机的针孔模型,基于张正友的平面标定方法,通过数字图像处理技术实现了对传送带上产品的准确定位.在定位过程中引入了图片加框技术,显著降低了复杂场景带来的误差.实验结果表明:该方法可以有效地实现传送带上单个或多个物体的定位;其中单物体定位的平均误差为1.8‰,多物体定位的平均误差小于1.1‰.
This paper presents a positioning technology for the products on conveying belt using monocular vision.Using the pinhole model,Zhang Zhengyou＇s camera calibration method and digital image processing technology,the conveying belt products positioning is realized.The picture framing technology is introduced in the process of positioning and it significantly reduces the error brought by complex scene.The experimental results show that this method can effectively achieve the single or multiple objects on conveying belt;the average error of single object positioning is 1.8‰;the average error of multiple objects positioning is less than
出处
《三峡大学学报（自然科学版）》
CAS
2015年第2期67-70,共4页
Journal of China Three Gorges University：Natural Sciences
基金
国家自然科学基金(11347117)
三峡大学优秀人才科研启动基金(KJ2011B066)
关键词
传送带
单目视觉
定位
数字图像处理
conveying belt
monocular vision
positioning
digital image processing
