摘要
该文基于TensorFlow开发平台,搭建了生成对抗网络模型,通过无人机遥感图像预处理、道路标签制作、网络训练和算法改进,完成了道路特征信息的深度学习,固化了生成网络参数,实现了基于无人机图像的低等级道路信息自动提取的科学目标,并通过形态学处理进一步增强了道路的提取效果。通过分析不同特征道路提取的信息,证明了本方法对利用高分辨率无人机图像提取低等级道路信息具有较好的借鉴作用。
Focused on scientific goals of automatic extraction of low-grade road information from unmanned aerials vehicle (UAV) images, the paper proposes a method based on generative adversarial networks (GANs). First, GANs are built based on TensorFlow platform;second, the UAV images are preprocessed and the road information is tagged in such images;third, the road feature of UAV images is trained by GANs built in the research. Finally, the low-grade road information from UAV images can be extracted automatically. Comparison and verification show that the proposed method provide a reference for extraction of road information.
作者
何磊
李玉霞
彭博
吴焕萍
HE Lei;LI Yu-xia;PENG Bo;WU Huan-ping(School of Software Engineering, Chengdu University of Information Technology Chengdu 610225;Automatic Software Generation and Intelligence Service Key Laboratory of Sichuan Province Chengdu 610225;School of Automation Engineering, University of Electronic Science and technology of China Chengdu 611731;Institute of Electronic and Information Engineering in Guangdong, University of Electronic Science and Techndogy of China Dongguan 523808;Operational System Development and Maintenance Division, National Climate Center Haidian Beijing 100081)
出处
《电子科技大学学报》
EI
CAS
CSCD
北大核心
2019年第4期580-585,共6页
Journal of University of Electronic Science and Technology of China
基金
四川省科技厅重点研发项目(2018SZ0286,2018GZ0099)
广东省自然科学基金(2018A030313898)
公益性行业(气象)科研专项(GYHY201506025)
关键词
生成对抗网络
图像处理
道路提取
无人机图像
generative adversarial network
image processing
road extraction
UAV image
