基于AR技术的光交箱资源校准技术  

Optical Box Resource Calibration Technology Based on AR Technology

作　　者：孔轶[1] 吴威[1] 张晓峰[1] KONG Yi;WU Wei;ZHANG Xiaofeng(China Mobile Group Guangdong Co.,Ltd.,Guangzhou 510000,China)

机构地区：[1]中国移动通信集团广东有限公司,广州510000

出　　处：《移动信息》2024年第1期156-159,共4页MOBILE INFORMATION

摘　　要：光交箱端口状态对通信网络的正常运行具有重要作用,由于光交箱内部布线杂乱,容易对端口产生排线遮挡问题,导致传统的光交箱端口图像识别操作不便。文中提出了一种基于AR技术的光交箱资源校准技术,该技术可以对实时视频自适应选取关键帧,然后对关键帧进行光交箱端口状态和文字识别,最后返回加权融合后的识别结果,并利用实时AR交互指引巡检人员处理被遮挡端口,这不仅能提升巡检人员的工作效率,还能提高实时识别准确率。将该校准技术应用到光交箱端口的智能检测中,能对光交箱的端口识别提供极大便利。The status of the optical junction box port plays an important role in the normal operation of the communication network.Due to the disorderly internal wiring of the optical junction box,it is easy to cause cable blocking problems on the port,which leads to inconvenience in traditional optical junction box port image recognition operations.This paper proposes a light box resource calibration technology based on AR technology,which can adaptively select keyframes for real-time videos,then recognize the port status and text of the light box for the keyframes,and finally return the weighted fusion recognition result.Real time AR interaction is used to guide inspectors to handle oceluded ports.This not only improves the work efficiency of inspectors,but also improves the real-time recognition accuracy.Applying this calibration technology to intelligent detection of optical junction box ports can provide great convenience for port recognition of optical junction boxes.

关 键 词：光交箱 AR交互 校准技术 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]