摘要
在煤炭工业生产运输过程中可能出现皮带打滑问题影响设备运行,现有的基于传感器的测速方法可能因为贴合问题出现误测。提出一种基于视觉的智能皮带测速方法,针对皮带低纹理特征点不足的问题,通过累积多帧特征点对估计运动模型。同时,实际场景可能出现的视频抖动问题可以通过背景特征点估计出偏差,并进行补偿。另外,图像坐标系下的检测结果可以根据标定得到的摄像机内外参数转换到世界坐标系下皮带运动速度。实验结果验证了方法的有效性和准确性。该方法实现了煤炭运输皮带的智能测速,可用于皮带打滑故障的及时报告。
In coal industry,a conveying belt may slip during coal transportation,thus affecting the efficient performance of the equipment.At present,conveyer speed detection methods depend on sensors,which may cause error due to the joint problem.Therefore,a vision-based smart speed detection approach was proposed.To handle the problem of insufficient featuring points in low-textured belt,motion model was estimated through accumulating point correspondences from multiple frames.Meanwhile,the video jitter problem could be eliminated using background feature points.In addition,the detection result in image coordinate system could be transformed to the actual speed of belt in a world coordination system by calibrating the internal and external parameters of camera.The experiment demonstrate the efficiency and accuracy of the proposed method,and realize smart speed measurement of conveyer belt and timely report of a slip problem.
作者
赵亮
韩宝虎
孙魁元
荣宝
屠凤招
张学慧
田立龙
张智浩
ZHAO Liang;HAN Bao-hu;SUN Kui-yuan;RONG Bao;TU Feng-zhao;ZHANG Xue-hui;TIAN Li-long;ZHANG Zhi-hao(Shenhua Baorixile Energy Co.,Ltd,Hulunbeier 021025,China;School of Astronautics,Harbin Institute of Technology,Harbin 150001,China)
出处
《科学技术与工程》
北大核心
2021年第12期4990-4994,共5页
Science Technology and Engineering
关键词
煤炭运输工业
皮带速度检测
特征点匹配
运动估计
摄像机标定
coal transportation industry
speed measurement of belt
feature point matching
motion estimation
camera calibration
