摘要
为得到垃圾填埋场渗滤液导排管道堵塞位置信息,提出基于相机标定的管道内壁堵塞位置检测方法。首先对导排管道堵塞图像进行中值滤波和图像增强处理;其次根据传统相机标定所涉及到的坐标系以及引入的新坐标系之间的关系建立矩阵方程模型,再利用畸变矫正对模型参数进行优化,通过极线矫正和图像匹配获取堵塞目标点在左右图像的像素坐标;最后代入矩阵方程模型求解堵塞位置与相机间距。实验结果表明:实际距离为2~5 m内时测距结果误差≤3.5 cm,满足工程要求。
In order to get the location information of landfill leachate drainage pipeline blockage,a pipeline inner wall blockage location detection method based on camera calibration was proposed in this paper.Firstly,median filtering and image enhancement were applied to the blockage image of the drainage pipeline.Secondly,the matrix equation model was established according to the relationship between the coordinate system involved in the traditional camera calibration and the introduced new coordinate system.Secondly,the model parameters were optimized by distortion correction,and the pixel coordinates of the blockage target point in the left and right images were obtained by epipolar correction and image matching.Finally,the matrix equation model was used to solve the relationship between blocking position and camera spacing.The experimental results show that when the actual distance is within 2~5 m,the ranging error is no more than 3.5 cm,which meets the engineering requirements.
作者
陈亚宇
王宏达
孙骥晟
李建龙
CHEN Ya-yu;WANG Hong-da;SUN Ji-sheng;LI Jian-long(School of Mechanical and Equipment Engineering,Hebei University of Technology,Handan 056038,China)
出处
《管道技术与设备》
CAS
2021年第4期6-10,共5页
Pipeline Technique and Equipment
基金
河北省省级科技计划资助(20373901D)。
关键词
导排管道
双目视觉
堵塞定位
图像处理
相机标定
leachate drainage pipeline
binocular vision
blockage positioning
image processing
camera calibration