电讯技术2018，Vol.58Issue(2)：119-125,7.DOI:10.3969/j.issn.1001-893x.2018.02.001
采用改进全卷积网络的"高分一号"影像居民地提取
Extraction of Residential Areas in GF-1 Remote Sensing Images Based on Improved Fully Convolutional Network
摘要
Abstract
The existing residential areas extraction methods suffer from low accuracy and efficiency. To o-vercome these problems,an improved fully convolutional network based residential areas extraction method is proposed. Firstly,large number of training samples are prepared by professional visual interpretation,and then a pre-trained deep convolution neural network is transformed into a fully convolutional network. Ad-ditionally,in order to reduce the amount of parameters and improve feature expression ability of the net-work,the convolutional layers transformed from fully connected layers are replaced with Inception module. Finally,the improved fully convolutional network is trained by the dataset prepared before. The experiment reveals that the proposed method achieves automatic,effective extraction of residential area information,and Kappa Coefficient is raised up to above 94%.关键词
"高分一号"卫星/高分辨率遥感图像/居民地信息提取/深度学习/全卷积网络Key words
GF-1 satellite/high resolution remote sensing image/residential areas extraction/deep learning/fully convolutional network分类
信息技术与安全科学引用本文复制引用
潘旭冉,杨帆,潘国峰..采用改进全卷积网络的"高分一号"影像居民地提取[J].电讯技术,2018,58(2):119-125,7.基金项目
国家科技重大专项(2009ZX02308-004) （2009ZX02308-004）
河北省高等学校科学研究项目(Z2014088) （Z2014088）
