摘要
为了避免带式输送机因负载超限而引发多种安全事故,提出了一种基于机器视觉的带式输送机负载超限检测方法。通过分析双目立体视觉成像原理,采用棋盘格标定方法结合OpenCV标定程序来完成双目相机的标定,利用灰度重心法来提取光条的中心线,得到煤料的点云数据,构建煤料轮廓的三维结构,并据此来完成输送机负载的计算。实验结果表明:重构后的三维图像能够较好地反映出煤料表面的起伏变化,与真实情况较为一致;检测精度和准确度较高,负载检测的绝对误差均小于1%;输送机负载发生超限时,界面能够弹出相应的警示。
In order to avoid a variety of safety accidents caused by overload of belt conveyor,a method for detecting overload of belt conveyor based on machine vision is presented.By analyzing the principle of binocular stereo vision imaging,the calibration of binocular camera is completed by checkerboard calibration method combined with OpenCV calibration program,the center line of light strip is extracted by gray centroid method,point cloud data of coal is obtained,three-dimensional structure of coal outline is constructed,and load calculation of conveyor is completed accordingly.The experimental results show that the reconstructed three-dimensional image can better reflect the undulating changes of coal surface,which is consistent with the real situation.The detection accuracy and accuracy are high,and the absolute error of load detection is less than 1%.When the loads of the conveyor exceed the limit,the interface can pop up the corresponding warning.
作者
林沣
LIN Feng(Guangxi Technological College of Machinery and Electricity,Nanning 530007,China)
出处
《煤炭技术》
CAS
北大核心
2023年第9期258-261,共4页
Coal Technology
基金
广西中青年教师基础能力提升项目(2022KY1071)。
关键词
机器视觉
灰度重心法
三维重构
负载超限
machine vision
gray centroid method
3D reconstruction
load overrun
