基于机器视觉和5G通信的输电线路异物检测方法  

A Foreign Object Detection Method for Transmission Lines Based on Machine Vision and 5G Communication

作　　者：王利[1] WANG Li(PrivateHualian College,Guangzhou Guangdong 510663 China)

机构地区：[1]私立华联学院,广东广州510663

出　　处：《长江信息通信》2025年第2期92-94,共3页Changjiang Information & Communications

摘　　要：为了准确识别和预警输电线路上的潜在异物,从而预防线路故障和安全事故,保障电网的稳定运行,提出一种基于机器视觉和5G通信的输电线路异物检测方法。通过5G通信技术实现输电线路图像的高速、实时传输,引入全局运动补偿机制,对采集得到的原始输电线路图像背景的位移量进行矫正处理,得到近似静态的输电线路图像背景信息;在具体的异物检测阶段,通过机器视觉算法中的Canny算子检测输电线路异物。在测试结果中,设计方法不仅对于输电线路异物检测结果的mAP值并未受到检测框尺度明显影响,对应的mAP值也始终稳定在较高水平,基本在94.0%以上,说明其检测效果较好。In order to accurately identify and warn potential foreign objects on transmission lines,prevent line faults and safety accidents,and ensure the stable operation of the power grid,a transmission line foreign object detection method based on machine vision and 5G communication is proposed.By using 5G communication technology to achieve high-speed and real-time transmission of transmission line images,a global motion compensation mechanism is introduced to correct the displacement of the background of the original transmission line image collected,and obtain approximately static background information of the transmission line image;In the specific foreign object detection stage,the Canny operator in machine vision algorithms is used to detect foreign objects in transmission lines.In the test results,the design method not only did not have a significant impact on the mAP value of the foreign object detection results of the transmission line due to the size of the detection box,but the corresponding mAP value also remained stable at a high level,basically above 94.0%,indicating that its detection effect is good.

关 键 词：机器视觉 5G通信 输电线路 异物检测 全局运动补偿 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]