摘要
针对传统桥吊的接触式摆角检测仪检测精确低、存在死区等缺点,且目前桥吊视觉检测方法的标识物过于繁琐且对旋转倾斜较为敏感的问题,提出了一种视觉目标检测算法来实时获取桥吊负载的空间位置。在桥吊负载上加球形标识物,通过获取球形标识物的信息来获得摆角与绳长信息。在设计出的搜索框中用改进的圆检测算法识别出球心位置与半径,之后通过空间几何方法来获取实时的摆角与绳长。利用桥吊视觉实验装置将提出的视觉检测技术与增量式编码器获取的结果进行对比分析,结果表明:视觉检测方法精确地获取摆角,为桥式吊车防摇控制系统提供更为精确的数据。
Aiming at the shortcomings such as low precision,dead zone exist in traditional contact angle encoders in overhead crane,and the marks used in visual detection systems are sensitive to rotation and tilt,a vision target detection algorithm is proposed to obtain the spatial position of the load in real time.Add a spherical marker to the load,the swing angle and rope length information by obtaining the information of the spherical marker.Firstly,the spherical center and the radius are identified by improved circle detection algorithm in the designed image processing area.Then,the space coordinates of spherical target are confirmed by the space geometry method.The experimental results indicate that the visual detection technology can provide more accurate data for control system of the overhead crane.
作者
张万鹏
徐为民
顾秀涛
张明明
王永爽
ZHANG Wanpeng;XU Weimin;GU Xiutao;ZHANG Mingming;WANG Yongshuang(Key Laboratory of Maritime Technology and Control Engineering of Ministry of Communications,Shanghai Maritime University,Shanghai 201306,China)
出处
《传感器与微系统》
CSCD
2020年第12期46-49,52,共5页
Transducer and Microsystem Technologies
基金
国家自然科学基金资助项目(16330492,15898654)。
关键词
视觉检测
霍夫变换
圆检测
机器视觉
桥式吊车
visual detection
Hough transform
circle detection
machine vision
bridge crane
