测绘通报 ›› 2019, Vol. 0 ›› Issue (7): 73-77.doi: 10.13474/j.cnki.11-2246.2019.0222

• 技术交流 • 上一篇    下一篇

利用全卷积网络提取Sentinel-2影像高低层建筑区

闫智1,2, 李利伟2, 程钢1   

  1. 1. 河南理工大学测绘与国土信息工程学院, 河南 焦作 454003;
    2. 中国科学院遥感与数字地球研究所中国科学院数字地球重点实验室, 北京 100094
  • 收稿日期:2018-10-29 出版日期:2019-07-25 发布日期:2019-07-31
  • 通讯作者: 李利伟。E-mail:lilw@radi.ac.cn E-mail:lilw@radi.ac.cn
  • 作者简介:闫智(1992-),男,硕士生,主要从事遥感图像智能解译研究。E-mail:993179344@qq.com
  • 基金资助:
    国家重点研发计划(2016YFB0501501);中科院战略性先导科技专项(XDA19080304);国家自然科学基金(91638201)

Extraction of high-rise and low-rise building areas from Sentinel-2 data based on full convolution networks

YAN Zhi1,2, LI Liwei2, CHENG Gang1   

  1. 1. School of Surveying and Land Information Engineering, Henan Polytechnic University, Jiaozuo 454003, China;
    2. Key Laboratory of Digital Earth Science, Chinese Academy of Sciences, Institute of Remote Sensing and Digital Earth, Chinese Academy of Sciences, Beijing 100094, China
  • Received:2018-10-29 Online:2019-07-25 Published:2019-07-31

摘要: 面向Sentinel-2多光谱数据,依据影像地物空间结构和地表实际功能类型将建筑区分为高层建筑区和低层建筑区,构建了一种基于全卷积网络模型的高、低层建筑区快速提取技术。在此基础上,以雄安新区及其周边为试验区,选取2017年3月获取的4景Sentinel-2多光谱数据进行试验验证和分析。结果表明:本文技术能够实现大范围区域内高层和低层建筑区的快速提取,总体平均提取精度达到95.30%,其中高层建筑区平均提取精度为99.22%,低层建筑区平均提取精度为91.38%,该技术明显优于现有基于纹理结构的高低层建筑区提取方法。通过对提取结果进行统计分析发现:约4.4×104 km2的研究区内高层和低层建筑区分别约为94和7351 km2;雄安新区中心三县内高层和低层建筑区分别约为1.25和312.24 km2。本文技术具有很好的推广性,结合Sentinel-2数据大幅宽高频次观测特点和更多类型建筑区样本,可以实现大范围地表多类型建筑区动态监测。

关键词: Sentinel-2, 全卷积网络, 高层, 低层, 建筑区, 提取

Abstract: This paper proposes a fully convolutional networks based method to intelligently exploit Sentinel-2 data for high-rise and low-rise building areas extraction. The building areas are divided into high-rise building areas and low-rise building areas according to their spatial structure in Sentinel-2 data and their real functional types. Four Sentinel-2 data covering the Xiong'an New Area and its surroundings in early 2017 is selected for experimental verification and analysis. The results show that the proposed algorithm can extract the high-rise and low-rise building areas from Sentinel-2 data in an effective and efficient manner. Overall accuracy of the two types of building areas is about 95.30%, of which the high-rise building areas accuracy is about 99.22%, and the accuracy of other building areas is about 91.38%. Compared with the existing texture-based method, the proposed method is more robust and fast. The high-rise and low-rise building areas in the study area covering about 44 000 km2 are about 94 and 7351 km2, respectively. The high-rise and low-rise building areas in the three core counties of Xiong'an New Area are about 1.25 and 312.24 km2, respectively. Our method can be easily extend to extract finer types of building areas given proper training samples of finer types of buildings areas and can realize large scale dynamic monitoring with the large imaging width and the high-frequency observation of Sentinel-2 data.

Key words: Sentinel-2, fully convolutional networks, high-rise, low-rise, building areas, extraction

中图分类号: