摘要
提出了一种基于卷积神经网络的智慧路灯联动控制算法,实现智慧路灯多功能的联动。为了提高联动的准确性,采用卷积神经网络对报警人和充电车辆进行检测。摄像头根据检测的结果自动调整角度,实时跟踪监测报警人和充电车辆的状态,并传回后台系统。通过实验证明了本文方法能提高联动的准确率,充分发挥各个功能模块的作用,极大地方便运维人员的日常工作。
A multi-functional linkage control algorithm of smart street lamp based on convolutional neural network( CNN) is presented in this paper. To enhance the linkage accuracy,CNN is used to detect alarming people and charging vehicle. The cameras automatically adjust angle according to the detection result,track alarming people and charging vehicle,and transfer the data to background system. The experiments validate that the linkage control algorithm can enhance the linkage accuracy,and achieve the full potential of smart street lamp functions. It is greatly convenient for the maintenance people.
作者
鄢小虎
李康
陈凯
YAN Xiaohu;LI Kang;CHEN Kai(School of Computer and Information Engineering,Hubei University,Wuhan 430062,China;State Grid Electric Power Research Institute,Wuhan 430074,China)
出处
《照明工程学报》
2018年第4期72-75,共4页
China Illuminating Engineering Journal
基金
湖北省自然科学基金--基于深度卷积神经网络的目标跟踪算法研究与应用(批准号:2017CFB305)
关键词
智慧路灯
卷积神经网络
联动控制
smart street lamp
convolutional neural network
linkage control