基于多源遥感的大尺度高分辨率不透水面深度学习提取方法

doi:10.11947/j.AGCS.2023.20210546

测绘学报 ›› 2023, Vol. 52 ›› Issue (2): 272-282.doi: 10.11947/j.AGCS.2023.20210546

基于多源遥感的大尺度高分辨率不透水面深度学习提取方法

孙根云^1,2, 王鑫¹, 安娜³, 张爱竹¹

1. 中国石油大学(华东)海洋与空间信息学院, 山东青岛 266580;
2. 海洋国家实验室海洋矿产资源评价与探测技术功能实验室, 山东青岛 266071;
3. 中国自然资源航空物探遥感中心, 北京 100083

收稿日期:2021-09-27 修回日期:2022-04-15 发布日期:2023-03-07
通讯作者: 安娜 E-mail:an_na826@163.com
作者简介:孙根云(1979-),男,教授,博士生导师,主要研究方向智能遥感、环境遥感。E-mail:genyunsun@163.com
基金资助:
国家自然科学基金（41971292）

A method for large-scale and high-resolution impervious surface extraction based on multi-source remote sensing and deep learning

SUN Genyun^1,2, WANG Xin¹, AN Na³, ZHANG Aizhu¹

1. College of Oceanography and Space Informatics, China University of Petroleum (East China), Qingdao 266580, China;
2. Laboratory for Marine Mineral Resources, Qingdao National Laboratory for Marine Science and Technology, Qingdao 266071, China;
3. China Aero Geophysical Survey and Remote Sensing Center for Natural Resources, Beijing 100083, China

Received:2021-09-27 Revised:2022-04-15 Published:2023-03-07
Supported by:
The National Natural Science Foundation of China (No. 41971292)

摘要/Abstract

摘要： 深度学习是提取不透水面的一类重要方法，具有精度高，泛化性强等优势。但是模型的训练需要依靠大量的训练样本。尤其是在高分辨率、大尺度不透水面制图时，获取数量足够且高质量的训练样本非常费时费力。因此，本文结合多源遥感影像与开源数据，提出了一种大尺度高分辨率不透水面自动提取方法。该方法首先从众源数据OpenStreetMap中自动获取训练样本，然后用开源的不透水面产品对噪声样本加权，减小标签噪声对模型训练的负面影响；在此基础上，构建了一种三分支的超轻量级CNN模型，融合光学、SAR和地形数据生成10 m不透水面产品。以越南全境为试验区，对本文方法进行了验证。试验结果表明，本文提出的方法分类总体精度和Kappa系数分别为91.01%和0.82，优于目前已发布的不透水面产品。本文研究成果可为澜湄流域等热带亚热带城市可持续发展和生态环境保护提供基础技术和数据支撑。

关键词: 不透水面, 深度学习, 众源数据, 卷积神经网络, 多源数据融合

Abstract: Deep learning is an important method for extracting impervious surfaces (IS), which has the advantages of high accuracy and generalization. However, the training of the models relies on a huge of train samples. Especially in large-scale and high-resolution IS mapping, it is time-consuming and laborious to obtain sufficient and high-quality training samples. Therefore, this study proposes a method to automatically extract IS based on multi-source remote sensing images and open-source data. Firstly, training samples are automatically obtained from crowdsourced OpenStreetMap data, and then the noise samples are weighted with open-source IS maps to reduce the negative influence of label noise on model training. Moreover, an ultra-lightweight CNN model with three branches was constructed to generate 10 m IS products by integrating optical, SAR and terrain data. In this paper, the method was validated in Vietnam. The results show that the overall accuracy and Kappa coefficient of the method proposed are 91.01% and 0.82, respectively, which are better than the currently released IS products. The research results of this paper can provide basic technology and data support for the sustainable development and ecological environment protection of tropical and subtropical cities in the Lancing-Mekong River basin.

Key words: impervious surface, deep learning, crowdsourced data, CNN, multisource data fusion

中图分类号:

P237

孙根云, 王鑫, 安娜, 张爱竹. 基于多源遥感的大尺度高分辨率不透水面深度学习提取方法[J]. 测绘学报, 2023, 52(2): 272-282.

SUN Genyun, WANG Xin, AN Na, ZHANG Aizhu. A method for large-scale and high-resolution impervious surface extraction based on multi-source remote sensing and deep learning[J]. Acta Geodaetica et Cartographica Sinica, 2023, 52(2): 272-282.

参考文献

[1] COHEN B. Urbanization in developing countries:current trends, future projections, and key challenges for sustainability[J]. Technology in Society, 2006, 28(1-2):63-80.
[2] BRABEC E, SCHULTE S, RICHARDS P L. Impervious surfaces and water quality:a review of current literature and its implications for watershed planning[J]. Journal of Planning Literature, 2002, 16(4):499-514.
[3] FEI Weicheng, ZHAO Shuqing. Urban land expansion in China's six megacities from 1978 to 2015[J]. Science of the Total Environment, 2019, 664:60-71.
[4] OPOKU A. Biodiversity and the built environment:implications for the sustainable development goals (SDGs)[J]. Resources, Conservation and Recycling, 2019, 141:1-7.
[5] GORELICK N, HANCHER M, DIXON M, et al. Google earth engine:planetary-scale geospatial analysis for everyone[J]. Remote Sensing of Environment, 2017, 202:18-27.
[6] TAMIMINIA H, SALEHI B, MAHDIANPARI M, et al. Google Earth Engine for geo-big data applications:a meta-analysis and systematic review[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2020, 164:152-170.
[7] CHEN Jun, CHEN Jin, LIAO Anping, et al. Global land cover mapping at 30 m resolution:a POK-based operational approach[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2015, 103:7-27.
[8] PESARESI M, EHRLICH D, FERRI S, et al. Operating procedure for the production of the global human settlement layer from Landsat data of the epochs 1975, 1990, 2000, and 2014[J]. Publications Office of the European Union, 2016:1-62.
[9] ZHANG Xiao, LIU Liangyun, WU Changshan, et al. Development of a global 30 m impervious surface map using multisource and multitemporal remote sensing datasets with the Google Earth Engine platform[J]. Earth System Science Data, 2020, 12(3):1625-1648.
[10] GONG Peng, WANG Jie, YU Le, et al. Finer resolution observation and monitoring of global land cover:first mapping results with Landsat TM and ETM+ data[J]. International Journal of Remote Sensing, 2013, 34(7):2607-2654.
[11] TU Ying, LANG Wei, YU Le, et al. Improved mapping results of 10 m resolution land cover classification in Guangdong, China using multisource remote sensing data with Google Earth Engine[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2020, 13:5384-5397.
[12] XU Rudong, LIU Jin, XU Jianhui. Extraction of high-precision urban impervious surfaces from Sentinel-2 multispectral imagery via modified linear spectral mixture analysis[J]. Sensors (Basel, Switzerland), 2018, 18(9):2873.
[13] SUN Genyun, CHEN Xiaolin, JIA Xiuping, et al. Combinational build-up index (CBI) for effective impervious surface mapping in urban areas[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2016, 9(5):2081-2092.
[14] 李益敏, 刘心知, 吴博闻, 等. 基于遥感技术的不透水面动态变化研究:以瑞丽市为例[J]. 云南大学学报(自然科学版), 2021, 43(4):716-724. LI Yimin, LIU Xinzhi, WU Bowen, et al. Dynamic change of impervious surface based on remote sensing technology:a case study of Ruili City[J]. Journal of Yunnan University (Natural Sciences Edition), 2021, 43(4):716-724.
[15] 李方刚, 李二珠, 阿里木·赛买提, 等. 融合多源时序遥感数据大尺度不透水面覆盖率估算[J]. 遥感学报, 2020, 24(10):1243-1254. LI Fanggang, LI Erzhu, SAMAT Alim, et al. Estimation of large-scale impervious surface percentage by fusion of multi-source time series remote sensing data[J]. Journal of Remote Sensing, 2020, 24(10):1243-1254.
[16] SUN Zhongchang, WANG Cuizhen, GUO Huadong, et al. A modified normalized difference impervious surface index (MNDISI) for automatic urban mapping from landsat imagery[J]. Remote Sensing, 2017, 9(9):942.
[17] 徐涵秋, 王美雅. 地表不透水面信息遥感的主要方法分析[J]. 遥感学报, 2016, 20(5):1270-1289. XU Hanqiu, WANG Meiya. Remote sensing-based retrieval of ground impervious surfaces[J]. Journal of Remote Sensing, 2016, 20(5):1270-1289.
[18] KRIZHEVSKY A, SUTSKEVER I, HINTON G E. ImageNet classification with deep convolutional neural networks[C]//Proceedings of the 25th International Conference on Neural Information Processing Systems. Lake Tahoe,NV,USA:ACM, 2012:1097-1105.
[19] SIMONYAN K, ZISSERMAN A. Very deep convolutional networks for large scale image recognition[C]//Proceedings of 2015 International Conference on Learning Representations. San Diego,CA,USA:ICLR, 2015:463-476.
[20] HE Kaiming, ZHANG Xiangyu, REN Shaoqing, et al. Deep residual learning for image recognition[C]//Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).Las Vegas, NV, USA:IEEE, 2016:770-778.
[21] GONG Jianya, JI Shunping. Photogrammetry and deep learning[J]. Journal of Geodesy and Geoinformation Science, 2018(1):1-15.
[22] 赵泉华, 谢凯浪, 王光辉, 等. 全卷积网络和条件随机场相结合的全极化SAR土地覆盖分类[J]. 测绘学报, 2020, 49(1):65-78. DOI:10.11947/j.AGCS.2020.20190038. ZHAO Quanhua, XIE Kailang, WANG Guanghui, et al. Land cover classification of polarimetric SAR with fully convolution network and conditional random field[J]. Acta Geodaetica et Cartographica Sinica, 2020, 49(1):65-78. DOI:10.11947/j.AGCS.2020.20190038.
[23] 张继贤, 顾海燕, 杨懿, 等. 高分辨率遥感影像智能解译研究进展与趋势[J]. 遥感学报, 2021, 25(11):2198-2210. ZHANG Jixian, GU Haiyan, YANG Yi, et al. Research progress and trend of high-resolution remote sensing imagery intelligent interpretation[J]. National Remote Sensing Bulletin, 2021, 25(11):2198-2210.
[24] HUANG Fenghua, YU Ying, FENG Tinghao. Automatic extraction of impervious surfaces from high resolution remote sensing images based on deep learning[J]. Journal of Visual Communication and Image Representation, 2019, 58:453-461.
[25] ZHANG Ce, PAN Xin, LI Huapeng, et al. A hybrid MLP-CNN classifier for very fine resolution remotely sensed image classification[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2018, 140:133-144.
[26] TONG Xinyi, XIA Guisong, LU Qikai, et al. Land-cover classification with high-resolution remote sensing images using transferable deep models[J]. Remote Sensing of Environment, 2020, 237:111322.
[27] DONG Runmin, LI Cong, FU Haohuan, et al. Improving 3-m resolution land cover mapping through efficient learning from an imperfect 10-m resolution map[J]. Remote Sensing, 2020, 12(9):1418.
[28] GONG Peng, LIU Han, ZHANG Meinan, et al. Stable classification with limited sample:transferring a 30-m resolution sample set collected in 2015 to mapping 10-m resolution global land cover in 2017[J]. Science Bulletin, 2019, 64(6):370-373.
[29] CORBANE C, SYRRIS V, SABO F, et al. Convolutional neural networks for global human settlements mapping from Sentinel-2 satellite imagery[J]. Neural Computing and Applications, 2021, 33(12):6697-6720.
[30] 林创. 面向遥感图像分类的卷积神经网络抗噪优化方法[D]. 深圳:中国科学院大学(中国科学院深圳先进技术研究院), 2021. LIN Chuang. Noise-resistant optimization method of convolutional neural network for remote sensing image classification[D]. Shenzhen:Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, 2021.
[31] FENG Shanshan, FAN Fenglei. A hierarchical extraction method of impervious surface based on NDVI thresholding integrated with multispectral and high-resolution remote sensing imageries[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2019, 12(5):1461-1470.

基于多源遥感的大尺度高分辨率不透水面深度学习提取方法

A method for large-scale and high-resolution impervious surface extraction based on multi-source remote sensing and deep learning

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	张永显, 马国锐, 崔志祥, 张志军. 面向大视角差的无人机影像序列学习型特征匹配[J]. 测绘学报, 2023, 52(2): 230-243.
[2]	卜丽静, 吴文玉, 张正鹏, 杨银. 先验引导的多降质特征“珞珈一号”夜光遥感影像质量提升方法[J]. 测绘学报, 2023, 52(2): 244-259.
[3]	禹文豪, 魏铖, 陈佳鑫. 面向不平衡POI类别分布的电子地图多层次标签预测模型[J]. 测绘学报, 2023, 52(2): 318-328.
[4]	张广斌, 高贤君, 冉树浩, 杨元维, 李丽珊, 张妍. 高分遥感影像云雪共存区轻量云高精度检测方法[J]. 测绘学报, 2023, 52(1): 93-107.
[5]	杨敏, 陈果, 李连营, 黄浩然, 苗静, 晏雄锋. 序列卷积神经网络支持下线状地图目标的分段方法[J]. 测绘学报, 2023, 52(1): 108-116.
[6]	眭海刚, 刘畅, 干哲, 江政杰, 徐川. 多模态遥感图像匹配方法综述[J]. 测绘学报, 2022, 51(9): 1848-1861.
[7]	张良培, 何江, 杨倩倩, 肖屹, 袁强强. 数据驱动的多源遥感信息融合研究进展[J]. 测绘学报, 2022, 51(7): 1317-1337.
[8]	程涛, 张洋, James Haworth. 基于网络和图的时空智能——概念、方法和应用[J]. 测绘学报, 2022, 51(7): 1629-1639.
[9]	龚健雅, 宦麟茜, 郑先伟. 影像解译中的深度学习可解释性分析方法[J]. 测绘学报, 2022, 51(6): 873-884.
[10]	张勤, 赵超英, 陈雪蓉. 多源遥感地质灾害早期识别技术进展与发展趋势[J]. 测绘学报, 2022, 51(6): 885-896.
[11]	龚健雅, 张觅, 胡翔云, 张展, 李彦胜, 姜良存. 智能遥感深度学习框架与模型设计[J]. 测绘学报, 2022, 51(4): 475-487.
[12]	毛文婧, 王卫林, 焦利民, 刘安宝. 基于深度学习的中国连续空间覆盖PM_2.5浓度预报[J]. 测绘学报, 2022, 51(3): 361-372.
[13]	金飞, 官恺, 刘智, 韩佳容, 芮杰, 李庆高. 无监督密集匹配特征提取网络性能分析[J]. 测绘学报, 2022, 51(3): 426-436.
[14]	何直蒙, 丁海勇, 安炳琪. 高分辨率遥感影像建筑物提取的空洞卷积E-Unet算法[J]. 测绘学报, 2022, 51(3): 457-467.
[15]	于洋洋, 贺康杰, 武芳, 许俊奎. 面状居民地形状分类的图卷积神经网络方法[J]. 测绘学报, 2022, 51(11): 2390-2402.