测绘学报 ›› 2016, Vol. 45 ›› Issue (10): 1210-1221.doi: 10.11947/j.AGCS.2016.20160183

• 摄影测量学与遥感 • 上一篇    下一篇

利用多光谱卫星遥感和深度学习方法进行青藏高原积雪判识

阚希1, 张永宏2,3, 曹庭2, 王剑庚4, 田伟5   

  1. 1. 南京信息工程大学大气科学学院, 江苏 南京 210044;
    2. 南京信息工程大学信息与控制学院, 江苏 南京 210044;
    3. 江苏省大气环境与装备技术协同创新中心, 江苏 南京 210044;
    4. 南京信息工程大学大气遥感学院, 江苏 南京 210044;
    5. 南京信息工程大学计算机与软件学院, 江苏 南京 210044
  • 收稿日期:2016-04-22 修回日期:2016-08-31 出版日期:2016-10-20 发布日期:2016-11-08
  • 通讯作者: 张永宏 E-mail:zyh@nuist.edu.cn
  • 作者简介:阚希(1987-),男,博士生,研究方向为深度学习与卫星遥感图像识别。E-mail:kanxi@nuist.edu.cn
  • 基金资助:

    国家自然科学基金(91337102;41401481);江苏省自然科学基金(BK20140997;14KJB170017)

Snow Cover Recognition for Qinghai-Tibetan Plateau Using Deep Learning and Multispectral Remote Sensing

KAN Xi1, ZHANG Yonghong2,3, CAO Ting2, WANG Jiangeng4, TIAN Wei5   

  1. 1. School of Atmospheric Science, Nanjing University of Information Science & Technology, Nanjing 210044, China;
    2. School of Information & Control, Nanjing University of Information Science & Technology, Nanjing 210044, China;
    3. Jiangsu Collaborative Innovation Center on Atmospheric Environment and Equipment Technology, Nanjing 210044, China;
    4. School of Atmospheric Physics, Nanjing University of Information Science & Technology, Nanjing 210044, China;
    5. School of Computer & Software, Nanjing University of Information Science & Technology, Nanjing 210044, China
  • Received:2016-04-22 Revised:2016-08-31 Online:2016-10-20 Published:2016-11-08
  • Supported by:

    The National Natural Science Foundation of China(Nos.91337102;41401481);Natural Science Foundation of Jiangsu Province (Nos.BK20140997;14KJB170017)

摘要:

青藏高原积雪对全球气候变化十分重要,针对已有积雪遥感判识方法中普遍采用的可见光与红外光谱数据易受复杂地形与高海拔影响,导致青藏高原地区积雪判识精度较低的问题,提出了一种基于多光谱遥感与地理信息数据特征级融合的积雪遥感判识方法:以风云三号卫星可见光与红外多光谱遥感资料与多要素地理信息作为数据源,由地面实测雪深数据与现有积雪产品交叉筛选出样本标签,构建并训练基于层叠去噪自编码器(SDAE)的特征融合与分类网络,从而有效辨识青藏高原遥感图像中的云、积雪以及无雪地表。经地面实测雪深数据验证,该方法分类精度显著高于使用相同数据源的FY-3A/MULSS积雪产品,略高于国际主流积雪产品MOD10A1与MYD10A1,并且年均云覆盖率最低。试验结果表明该方法可有效地减少云层对积雪判识的干扰,提升分类精度。

关键词: 卫星遥感, 深度学习, 风云三号, 多光谱融合, 青藏高原, 积雪

Abstract:

Snow cover in Qinghai-Tibetan plateau (QT plateau) is very important to global climate change. Because of the complex topography and high altitude, the recognition accuracies of existing snow cover products in QT plateau are significantly lower than flat areas. This paper proposed a new method of snow cover recognition for QT plateau based on deep learning. The multispectral remote sensing data from Chinese meteorological satellite FY-3A and the multiple geographic elements information are put together as the data sources, the insitu snow depth measurements and existing snow cover products are used for selecting the labeled samples. A stacked denoising auto-encoders (SDAE) network was built and trained for feature extraction and classification, this network can be used as a classifier for distinguishing the snow cover from cloud and other snow-free surface features. The recognition results are verified by snow depth data of meteorological station observations, verification results show that the recognition accuracy of this method is significantly higher than the snow product FY-3A/MULSS, which is using the same remote sensing data source FY-3A, and slightly higher than the widely used snow products MOD10A1 and MYD10A1,and the cloud coverage rate of this method is the lowest. According to the validation results, this method can effectively improve the accuracy of snow cover recognition, and reduce the interference of clouds.

Key words: satellite remote sensing, deep learning, FengYun-3, multispectral data fusion, Qinghai-Tibetan plateau, snow cover

中图分类号: