摘要
螺母作为输电线路上的重要连接构件,对于保证输电安全至关重要。文章设计了基于改进深度霍夫的螺母中心定位检测方法,将目标检测技术引入到输电杆塔等现场检测单元中。将工业相机采集到的工件图像输入到YOLOv5检测网络中得出目标框中心坐标,采用Zero-DCE图像增强算法进行预处理,最后通过深度霍夫直线检测与K-means聚类算法得出螺母边缘直线,通过六边形约束法找到螺母中心点坐标,将位置坐标通过串口传输到上位机控制机械臂紧固螺母。实验结果表明,所提算法能够实现螺母中心定位,具有较强适用性和鲁棒性。
As an important connecting component of transmission line,nut is very important to ensure the safety of transmission.In this paper,a nut center detection method based on improved depth Hough is designed,and the object detection technology is introduced into the field detection units such as transmission towers.The workpiece image collected by the industrial camera is input into the YOLOv5 detection network to obtain the center coordinates of the target frame.The zero DCE image enhancement algorithm is used for preprocessing.Finally,the nut edge line is obtained through the deep Hough line detection and K-means clustering algorithm.The nut center point coordinates are found through the hexagon constraint method,and the position coordinates are transmitted to the upper computer through the serial port to control the mechanical arm to tighten the nut.The experimental results show that the proposed algorithm can realize the nut center positioning,and has strong applicability and robustness.
作者
翟永杰
张效铭
白云山
王乾铭
李冰
ZHAI Yongjie;ZHANG Xiaoming;BAI Yunshan;WANG Qianming;LI Bing(Department of Automation,North China Electric Power University,Baoding 071000,China)
出处
《电力信息与通信技术》
2022年第11期13-19,共7页
Electric Power Information and Communication Technology
基金
国家自然科学基金联合基金项目重点支持项目“可视度受限环境下跨光谱多传感信息融合的机器人语义感知与交互协作”(U21A20486)。
