Please wait a minute...
img

官方微信

遥感技术与应用  2020, Vol. 35 Issue (6): 1329-1336    DOI: 10.11873/j.issn.1004-0323.2020.6.1329
冰雪遥感专栏     
基于高分遥感数据和深度学习的石冰川自动提取研究
徐瑾昊1,2(),冯敏2,3(),王建邦2,3,冉有华4,祁元4,5,杨联安1,李新2
1.西北大学城市与环境学院,陕西 西安 710127
2.中国科学院青藏高原研究所 三极监测与大数据中心,北京 100101
3.兰州大学资源环境学院,甘肃 兰州 730000
4.中国科学院西北生态环境资源研究院,甘肃 兰州 730000
5.高分辨率对地观测系统甘肃数据与应用中心,甘肃 兰州 730000
Automatically Identifying Rock Glacier based on Gaofen Satellite Image and Deep Learning
Jinhao Xu1,2(),Min Feng2,3(),Jianbang Wang2,3,Youhua Ran4,Yuan Qi4,5,Lian’an Yang1,Xin Li2
1.College of Urban and Environmental Sicence,Northwest University,Xi’an 710127,China
2.Institude of Tibetan Plateau Research,Chinese Academy of Sciences,Beijing,Beijing 100101,China
3.College of Earth and Environmental Sciences,Lanzhou University,Lanzhou 730000,China
4.Northwest Institute of Ecology and Environmental Resources,Chinese Academy of Sciences,Lanzhou 730000,China
5.Data and Application Center of High-Resolution Earth Observation System in Gansu,Lanzhou 730000,China
 全文: PDF(9437 KB)   HTML
摘要:

石冰川是以冰岩混合物为基础形成的一类具有舌状堆积纹理的冰缘地貌,了解其分布和变化对于寒区环境研究具有重要价值,遥感技术的发展为石冰川的识别提供了有效的手段。针对石冰川发育地的偏远和调查的困难,以及其光谱特征的微弱性,提出了一种基于深度学习的石冰川识别方法,以ResNet作为训练网络,得到石冰川的图像分类模型,以国产高分一号遥感影像作为实验数据,在念青唐古拉山西段展开了应用,共识别出石冰川96条。验证结果表明:该方法具有较高的识别精度(98.72%的总体精度、89.48%的生产精度和81.77 %的用户精度),证明该方法能够有效地识别石冰川,并为在大区域开展石冰川的调查和分析提供了基础。

关键词: 深度学习高分一号石冰川ResNet    
Abstract:

Rock glacier is a geomorphological landform with ligule accumulation texture that formed from mixture of ice and debris. Investigating the distribution of rock glaciers can provide effective information for studying the environment and climate change in cold regions. The development of remote sensing technology provides an effective way for identifying rock glacier. However, its execution is difficult due to the large area of rock glacier distribution as well as the similarity between rock glacier and its surroundings in spectral surface reflectance. Comparing to the traditional visual interpretation approach, this paper presented a more effective method for automatically identifying rock glaciers in high-resolution images. The method was implemented by integrating the Deep Learning development framework to build the model through interactive training from the ResNet network. The model was then applied to identify rock glaciers in GaoFen-1 images that collected in West Nyainqentanglha Mountains, where are rich of rock glaciers. Gaofen-1 images were used as the satellite data source, and 96 rock glaciers were identified in the West Nyainqentanglha Mountains. Accuracy of the results were assessed by comparing to human interpreted data, and it reported 98.72% Overall Accuracy, 89.48% Producer's Accuracy, and 81.77% User's Accuracy, suggesting that the presented method is very effective for identifying rock glaciers, and it provides a potential capability for mapping the distribution of rock glacier in large areas.

Key words: Deep Learning    Gaofen-1    Rock Glacier    ResNet
收稿日期: 2019-10-20 出版日期: 2021-01-26
ZTFLH:  TP753  
基金资助: 中国科学院战略性先导科技专项(XDA20100104);中国科学院百人计划资助
通讯作者: 冯敏     E-mail: viktorxu@qq.com;mfeng@itpcas.ac.cn
作者简介: 徐瑾昊(1997-),男,湖南常德人,硕士研究生,主要从事深度学习在遥感方面的应用。E?mail:viktorxu@qq.com
服务  
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章  
徐瑾昊
冯敏
王建邦
冉有华
祁元
杨联安
李新

引用本文:

徐瑾昊,冯敏,王建邦,冉有华,祁元,杨联安,李新. 基于高分遥感数据和深度学习的石冰川自动提取研究[J]. 遥感技术与应用, 2020, 35(6): 1329-1336.

Jinhao Xu,Min Feng,Jianbang Wang,Youhua Ran,Yuan Qi,Lian’an Yang,Xin Li. Automatically Identifying Rock Glacier based on Gaofen Satellite Image and Deep Learning. Remote Sensing Technology and Application, 2020, 35(6): 1329-1336.

链接本文:

http://www.rsta.ac.cn/CN/10.11873/j.issn.1004-0323.2020.6.1329        http://www.rsta.ac.cn/CN/Y2020/V35/I6/1329

图1  念青唐古拉山西段
图2  方法流程图
图3  训练数据样本
图4  CNN网络结构
图5  念青唐古拉山西段石冰川空间分布
图6  典型石冰川识别效果
模型总体精度/%生产精度/%用户精度/%
均值98.7289.4881.77
199.1890.6389.69
298.4590.6376.32
398.8487.5084.85
498.7589.5881.90
598.5888.5479.44
698.1187.5073.04
799.0187.5088.42
898.9793.7583.33
998.7591.6780.73
1098.5887.5080.00
表1  石冰川分类模型精度评价
1 Qin Dahe, Yao Tandong, Ding Yongjian, et al. Glossary of Cryosphere Science (Revised)[M]. Beijing: China Meteorological Press, 2016.秦大河, 姚檀栋, 丁永建,等. 冰冻圈科学辞典(修订版)[M]. 北京: 气象出版社, 2016.
2 Giardino J R, Vick S G. Geologic Engineering Aspects of Rock Glaciers[M]. Bonston:Allen and Unwin, 1987: 265-287.
3 Humlum O. The Climatic Significance of Rock Glaciers[J]. Permafrost and Periglacial Processes, 1998, 9(4): 375-395.
4 Konrad S K, Humphrey N F, Steig E J, et al. Rock Glacier Dynamics and Paleoclimatic Implications[J]. Geology, 1999, 27(12): 1131-1134.
5 Millar C I, Westfall R D, Evenden A, et al. Potential Climatic Refugia in Semi-Arid, Temperate Mountains: Plant and Arthropod Assemblages Associated with Rock Glaciers, Talus Slopes, and Their Forefield Wetlands, Sierra Nevada, California, USA[J]. Quaternary International, 2015, 387: 106-121. doi: 10.1016/j.quaint.2013.11.003.
doi: 10.1016/j.quaint.2013.11.003
6 Brenning A. Geomorphological, Hydrological and Climatic Significance of Rock Glaciers in the Andes of Central Chile (33-35 S)[J]. Permafrost and Periglacial Processes, 2005, 16(3): 231-240. doi: 10.1002/ppp.528.
doi: 10.1002/ppp.528
7 Janke J R, Bellisario A C, Ferrando F A. Classification of Debris-covered Glaciers and Rock Glaciers in the Andes of Central Chile[J]. Geomorphology, 2015, 241: 98-121. doi: 10.1016/j.geomorph.2015.03.034.
doi: 10.1016/j.geomorph.2015.03.034
8 Lugon R, Stoffel M. Rock-glacier Dynamics and Magnitude-frequency Relations of Debris Flows in a High-elevation Watershed: Ritigraben,Swiss Alps[J]. Global and Planetary Change, 2010,73(3-4):202-210. doi: 10.1016/j.gloplacha. 2010.06.004.
doi: 10.1016/j.gloplacha. 2010.06.004
9 Cui Zhijiu. Discovery of KunLunShan-type Rock Glaciers and the Classification of Rock Glaciers[J]. Chinese Science Bulletin, 1984, 29(13): 810-813.
9 崔之久. 昆仑山型石冰川的发现及石冰川的最新分类[J]. 科学通报, 1984, 29(13): 810-813.
10 Cui Zhijiu, Zhu Cheng. The Structural Type of Temperature and Mechanism of Movement of Rock Glacier at the Head of Urumqi River, Tianshan Mountains[J]. Chinese Science Bulletin, 1989, 34(2): 134-137.
10 崔之久, 朱诚. 天山乌鲁木齐河源区石冰川的温度结构类型与运动机制[J]. 科学通报, 1989, 34(2): 134-137.
11 Li Shude, Yao Heqing. Preliminary Study of the Rock Glaciers in the Gongga Mt. Area[J]. Journal of Glaciology and Geocryology, 1987, 9(1): 55-60, 54.
11 李树德, 姚河清. 贡嘎山山地石冰川的初步研究[J]. 冰川冻土, 1987, 9(1): 55-60,54.
12 Liu Gengnian, Xiong Heigang, Cui Zhijiu, et al. The Morphological Features and Environmental Condition of RockGlaciers in TianShan Mountains[J]. Scientia Geographica Sinica, 1995, 15(3): 226-233,297.
12 刘耕年, 熊黑钢, 崔之久,等. 天山石冰川的形态与发育条件[J]. 地理科学, 1995, 15(3): 226-233,297.
13 Zhu Cheng, Cui Zhijiu, Yao Zeng. Research of the Feature of Rock Glacier on the Central Tian Shan Mountains[J]. Acta Geographica Sinaca, 1992(3): 233-241,289-290.
13 朱诚, 崔之久, 姚增. 中天山石冰川特征研究[J]. 地理学报, 1992(3): 233-241,289-290.
14 Blöthe J H, Rosenwinkel S, Höser T, et al. Rock-glacier Dams in High Asia[J]. Earth Surface Processes and Landforms, 2019, 44(3): 808-824. doi: 10.1002/esp.4532.
doi: 10.1002/esp.4532
15 Pandey P. Inventory of Rock Glaciers in Himachal Himalaya, India Using High-resolution Google Earth Imagery[J]. Geomorphology, 2019, 340: 103-115. doi: 10.1016/j.geomorph.2019.05.001.
doi: 10.1016/j.geomorph.2019.05.001
16 Ran Z, Liu G. Rock Glaciers in Daxue Shan, South-Eastern Tibetan Plateau: An Inventory, Their Distribution, and Their Environmental Controls[J]. Cryosphere, 2018, 12(7): 2327-2340. doi: 10.5194/tc-12-2327-2018.
doi: 10.5194/tc-12-2327-2018
17 Wang X, Liu L, Zhao L, et al. Mapping and Inventorying Active Rock Glaciers in the Northern Tien Shan of China Using Satellite SAR Interferometry.[J]. Cryosphere, 2017, 11(2): 997-1014. doi: 10.5194/tc-11-997-2017.
doi: 10.5194/tc-11-997-2017
18 Xu Junli, Liu Shiyin, Wang Jian. Distribution Characteristics of Rock Glaciers in the Upstream of Bayu Hydropower Station in Sangri County,Tibet[J]. Journal of Glaciology and Geocryology, 2018, 40(6): 1207-1215.
18 许君利, 刘时银, 王建. 西藏桑日县巴玉水电站上游石冰川分布特征[J]. 冰川冻土, 2018, 40(6): 1207-1215.
19 Krizhevsky A, Sutskever I, Hinton G E. Imagenet Classification with Deep Convolutional Neural Networks[C]//Advances in Neural Information Processing Systems. 2012: 1097-1105. doi: 10.1145/3065386.
doi: 10.1145/3065386
20 Geirhos R, Rubisch P, Michaelis C, et al. ImageNet-trained CNNs are Biased Towards Texture; Increasing Shape Bias Improves Accuracy and Robustness[J]. ArXiv Preprint ArXiv:1811.12231, 2018.
21 Hu F, Xia G S, Hu J, et al. Transferring Deep Convolutional Neural Networks for the Scene Classification of High-resolution Remote Sensing Imagery[J]. Remote Sensing, 2015, 7(11): 14680-14707. doi: 10.3390/rs71114680.
doi: 10.3390/rs71114680
22 Cheng G, Zhou P, Han J. Learning Rotation-invariant Convolutional Neural Networks for Object Detection in VHR Optical Remote Sensing Images[J]. IEEE Transactions on Geoscience and Remote Sensing, 2016, 54(12): 7405-7415. doi: 10.1109/tgrs.2016.2601622.
doi: 10.1109/tgrs.2016.2601622
23 Maggiori E, Tarabalka Y, Charpiat G, et al. Convolutional Neural Networks for Large-scale Remote-sensing Image Classification[J]. IEEE Transactions on Geoscience and Remote Sensing,2016, 55(2): 645-657. doi: 10.1109/tgrs.2016. 2612821.
doi: 10.1109/tgrs.2016. 2612821
24 Liu W, Guo Q, Wang Y. Temporal-spatial Climate Change in the Last 35 Years in Tibet and Its Geo-Environmental Consequences[J]. Environmental Geology, 2008, 54(8): 1747-1754. doi: 10.1007/s00254-007-0952-y.
doi: 10.1007/s00254-007-0952-y
25 Rees H G, Collins D N. Regional Differences in Response of Flow in Glacier-fed Himalayan Rivers to Climatic Warming[J]. Hydrological Processes: An International Journal, Wiley Online Library, 2006, 20(10): 2157-2169. doi: 10.1002/hyp.6209.
doi: 10.1002/hyp.6209
26 Bai Zhaoguang. Technical Features of Gaofen-1 Satellite[J]. Aerospace China, 2013(8): 5-9.白照广. 高分一号卫星的技术特点[J]. 中国航天, 2013(8): 5-9.
27 An Guoying, Han Lei, Huang Shuchun, et al. Dynamic Variation of Glaciers in Nyainqentanglha Mountain during 1999-2015: Evidence from Remote Sensing[J]. Geosience, 2019, 33(1): 176-186.
27 安国英, 韩磊, 黄树春,等. 念青唐古拉山现代冰川1999-2015年期间动态变化遥感研究[J]. 现代地质, 2019, 33(1): 176-186.
28 Xie Jian, Liu Jingshi, Du Mingyuan, et al. Hydrothermal Characteristics of the Land-atmospheric System in an Alpine Area of West Nyainqentanglha Mountains. Progress in Geography, 2010, 29(2): 151-158.[谢健, 刘景时, 杜明远,等. 念青唐古拉山西段高海拔陆—气系统水热特征[J]. 地理科学进展, 2010, 29(2): 151-158.]
29 He K, Zhang X, Ren S, et al. Deep Residual Learning for Image Recognition[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2016: 770-778. doi: 10.1109/cvpr.2016.90.
doi: 10.1109/cvpr.2016.90
[1] 杨瑞,祁元,苏阳. 深度学习U-Net方法及其在高分辨卫星影像分类中的应用[J]. 遥感技术与应用, 2020, 35(4): 767-774.
[2] 张峰极,吴艳兰,姚雪东,梁泽毓. 基于改进DenseNet网络的多源遥感影像露天开采区智能提取方法[J]. 遥感技术与应用, 2020, 35(3): 673-684.
[3] 程春梅,韦玉春,李渊,涂乾光. 太湖水体GF-1/WFV影像的6S逐像元大气校正[J]. 遥感技术与应用, 2020, 35(1): 141-152.
[4] 李越帅,郑宏伟,罗格平,杨辽,王伟胜,桂东伟. 集成U-Net方法的无人机影像胡杨树冠提取和计数[J]. 遥感技术与应用, 2019, 34(5): 939-949.
[5] 孙珊珊,田昕,谷成燕,韩宗涛,王崇阳,张兆鹏. 基于KNN-FIFS的内蒙古根河森林郁闭度遥感估测研究[J]. 遥感技术与应用, 2019, 34(5): 959-969.
[6] 崔刚,吴金胜,于镇,周玲. 深度学习对不同分辨率影像冬小麦识别的适用性研究[J]. 遥感技术与应用, 2019, 34(4): 748-755.
[7] 王宇,杨艺,王宝山,王田,卜旭辉,王传云. 深度残差神经网络高分辨率遥感图像建筑物分割[J]. 遥感技术与应用, 2019, 34(4): 736-747.
[8] 王颖洁,张荞,张艳梅,蒙印,郭文. 基于深度卷积神经网络的油罐目标检测研究[J]. 遥感技术与应用, 2019, 34(4): 727-735.
[9] 林志玮,丁启禄,黄嘉航,涂伟豪,胡典,刘金福. 基于DenseNet的无人机光学图像树种分类研究[J]. 遥感技术与应用, 2019, 34(4): 704-711.
[10] 刘天福,陈学泓,董琪,曹鑫,陈晋. 深度学习在GlobeLand30-2010产品分类精度优化中应用研究[J]. 遥感技术与应用, 2019, 34(4): 685-693.
[11] 徐梦竹, 徐佳, 邓鸿儒, 袁春琦. 基于全极化SAR影像的海岛地物分类[J]. 遥感技术与应用, 2019, 34(3): 647-654.
[12] 李淑敏, 冯权泷, 梁其椿, 张学庆. 基于深度学习的国产高分遥感影像飞机目标自动检测[J]. 遥感技术与应用, 2018, 33(6): 1095-1102.
[13] 孙红, 田昕, 闫敏, 李增元, 陈尔学, 孙珊珊, 王崇阳. 内蒙古大兴安岭根河植被覆盖度动态变化及影响因素的分析[J]. 遥感技术与应用, 2018, 33(6): 1159-1169.
[14] 江东,陈帅,丁方宇,付晶莹,郝蒙蒙. 基于面向对象的遥感影像分类研究——以河北省柏乡县为例[J]. 遥感技术与应用, 2018, 33(1): 143-150.
[15] 何海清,庞燕,陈晓勇. 面向遥感影像场景的深度卷积神经网络递归识别模型[J]. 遥感技术与应用, 2017, 32(6): 1078-1082.