液晶与显示2018，Vol.33Issue(4)：317-325,9.DOI:10.3788/YJYXS20183304.0317
基于深度卷积神经网络的输电线路可见光图像目标检测
Object detection of transmission line visual images based on deep convolutional neural network
周筑博 1高佼 2张巍 3王晓婧 1张静1
作者信息
- 1. 天津航天中为数据系统科技有限公司(天津市智能遥感信息处理技术企业重点实验室),天津 300301
- 2. 济南汤尼机器人科技有限公司,山东济南 250101
- 3. 南方电网科学研究院有限责任公司,广东广州 510080
- 折叠
摘要
Abstract
A deep convolutional neural network based method is adopted to detect objects such as tow-er,glass insulator and composite insulator in visible images of transmission lines.About 600 visible images of 19 different transmission lines are captured by manned helicopter with high-definition cam-era.All of the images are then annotated manually and segmented into blocks with 4 different labels:background,tower,glass insulator and composite insulator.These blocks are then augmented to around 150000 training samples which comprise the transmission line image dataset.A five-layer deep convolutional neural network is designed and pre-trained by using Cifar-100 dataset,the trained net-work is then fine-tuned by using transmission line image dataset.The experimental results show that when detection true positive rate is 90%,the false alarm rate is less than 10%,which is obviously su-perior to the traditional methods.It can be used for the detection of tower,glass insulator and com-posite insulator in visible images of transmission lines.T he detection result can be used as reference for diagnosis or state analysis of transmission lines.This method can be used to detect tower and insu-lator in visible images of transmission lines,and can be extended to detect other typical objects.关键词
输电线路图像/绝缘子/目标检测/深度学习/卷积神经网络Key words
transmission line image/insulator/object detection/deep learning/convolutional neural networks分类
信息技术与安全科学引用本文复制引用
周筑博,高佼,张巍,王晓婧,张静..基于深度卷积神经网络的输电线路可见光图像目标检测[J].液晶与显示,2018,33(4):317-325,9.
