Please wait a minute...
img

官方微信

遥感技术与应用  2023, Vol. 38 Issue (2): 319-331    DOI: 10.11873/j.issn.1004-0323.2023.2.0319
LUCC专栏     
基于LightGBM和多光谱—SAR多特征综合的中亚干旱区城市不透水面提取
刘希鸣1,2,3(),阿里木·赛买提1,2,3(),王伟1,2,3,吉力力·阿不都外力1,2,3
1.中国科学院新疆生态与地理研究所荒漠与绿洲生态国家重点实验室,新疆 乌鲁木齐 830011
2.中国科学院大学,北京 100049
3.中国科学院中亚生态与环境研究中心,新疆 乌鲁木齐 830011
LightGBM based Impervious Surface Area Extraction of Cities from Arid Areas in Central Asia Using Synthesized Multi-features of Multispectral-SAR Images
Ximing LIU1,2,3(),Alim SAMAT1,2,3(),Wei WANG1,2,3,Jilili ABUDUWAILI1,2,3
1.State Key Laboratory of Desert and Oasis Ecology,Xinjiang Institute of Ecology and Geography,Chinese Academy of Sciences,Urumqi 830011,China
2.University of Chinese Academy of Sciences,Beijing 100049,China
3.Research Center for Ecology and Environment of Central Asia,Chinese Academy of Sciences,Urumqi 830011,China
 全文: PDF(9727 KB)   HTML
摘要:

不透水面是评价城市化水平和城市生态环境的重要指标,是近年来城市遥感研究中的热点方向之一。与湿润、半湿润区相比,干旱区城市植被覆盖度较低,不透水面与裸土、荒漠之间相似的光谱特征导致传统基于光学影像的亚像元分解法与光谱指数法在干旱区不透水面提取的适用性降低。针对该问题,提出一种多光谱与合成孔径雷达(SAR)影像多特征综合的方法以增大不透水面与其他地物覆盖类型之间的特征差异,从而提取干旱区城市不透水面。以阿斯塔纳、塔什干和杜尚别3个中亚城市为研究区,哨兵2号和哨兵1号影像为数据源,通过LightGBM算法对多光谱和SAR图像的空间特征、SAR的极化特征进行分类并提取不透水面。研究对比了不同特征组合以及不同分类方法的不透水面提取结果,实验结果表明:多光谱与SAR影像多特征综合的方法能有效提高干旱区不透水面提取精度,明显改善干旱区其他土地覆盖类型错分为不透水面的问题;LightGBM算法与XGBoost、HistGBT等基于梯度提升决策树的算法和随机森林等方法相比能获取更高的精度,更适用于干旱区不透水面提取。这表明基于LightGBM以及多光谱和SAR多特征联合的方法能够有效提取中亚干旱区城市不透水面。

关键词: 不透水面部分重建的形态学属性剖面LightGBM干旱区中亚    
Abstract:

Impervious surface is an important factor indicates the level of urbanization and the urban ecological environment, and it is one of the current research hotspots in urban remote sensing. Compared with humid and semi-humid areas, urban vegetation coverage in arid areas is relatively low, the similar spectum between impervious surface and barren area makes the traditional optical image-based spectral mixing analysis method and spectral index method not suitable for the impervious surface extraction of cities in arid areas. In response to this problem, a method for impervious surfaces extraction of cities from arid areas in Central Asia using synthesized multi-features of multispectral-SAR images is proposed to improve the mixclassifiation between impervious surfaces and bare soil, so as to extract impervious surface in arid area. In detials, Sentinel-2 and the dual-polarization SAR image of Sentinel-1 are selected for three Central Asia cities, Astana, Tashkent and Dushanbe. The spatial characteristics of multi-spectral and SAR images, and the polarization characteristics of SAR are feeded to LightGBM algorithm to classify and extract impervious surface. This paper compares the impervious surface extraction results of different feature combinations and different classification methods. Experimental results indicated that the multi-feature synthesis method of multispectral and SAR images proposed can effectively improve the accuracy of impervious surface extraction in arid areas, indicating the improvement in the misclassification of impervious surface and other land cover types in arid areas; the LightGBM algorithm has higher accuracy than XGBoost, HistGBT and other algorithms based on gradient boosting decision trees and random forest algorithm, and it is more suitable for extraction of impervious surface in arid area. This shows that the method based on LightGBM and the combination of multispectral and SAR multi-features can effectively extract the urban impervious surface in the arid area of Central Asia.

Key words: Impervious surface    Morphological attribute profiles with partial reconstruction    LightGBM    Arid region    Central Asia
收稿日期: 2021-11-09 出版日期: 2023-05-29
ZTFLH:  P237  
基金资助: 国家自然科学基金面上项目“样本与特征迁移的中亚典型城市覆被精细分类方法研究”(42071424);中国科学院青年创新促进会项目(2018476)
通讯作者: 阿里木·赛买提     E-mail: liuximing20@mails.ucas.ac.cn;alim_smt@ms.xjb.ac.cn
作者简介: 刘希鸣(1998-),女,河南洛阳人,硕士研究生,主要从事城市遥感方面的研究。E?mail:liuximing20@mails.ucas.ac.cn
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  
刘希鸣
阿里木·赛买提
王伟
吉力力·阿不都外力

引用本文:

刘希鸣,阿里木·赛买提,王伟,吉力力·阿不都外力. 基于LightGBM和多光谱—SAR多特征综合的中亚干旱区城市不透水面提取[J]. 遥感技术与应用, 2023, 38(2): 319-331.

Ximing LIU,Alim SAMAT,Wei WANG,Jilili ABUDUWAILI. LightGBM based Impervious Surface Area Extraction of Cities from Arid Areas in Central Asia Using Synthesized Multi-features of Multispectral-SAR Images. Remote Sensing Technology and Application, 2023, 38(2): 319-331.

链接本文:

http://www.rsta.ac.cn/CN/10.11873/j.issn.1004-0323.2023.2.0319        http://www.rsta.ac.cn/CN/Y2023/V38/I2/319

图1  研究区地理位置示意图审图号:GS(2016)2938
波段中心波长/μm空间分辨率/m
B10.44360
B20.49010
B30.56010
B40.66510
B50.70520
B60.74020
B70.78320
B80.84210
B8a0.86520
B90.94560
B101.37560
B111.61020
B122.19020
表1  哨兵2号MSI数据介绍
城市成像时间数据类型
阿斯塔纳2020-09-20哨兵1号 GRD/SLC极化方式:VV-VH
2020-09-11哨兵2号 波段:B2、B3、B4、B8
塔什干2020-08-09哨兵1号 GRD/SLC极化方式:VV-VH
2020-08-04哨兵2号 波段:B2、B3、B4、B8
杜尚别2020-08-09哨兵1号 GRD/SLC极化方式:VV-VH
2020-08-07哨兵2号 波段:B2、B3、B4、B8
表 2  遥感影像数据表
图2  技术流程图
参数含义用途
max_depth树的最大深度合理的设置可以防止过拟合
num_leaves叶子节点数调节模型复杂度
learning_rate学习率一般学习率越高,模型收敛越快
lambda正则化参数一般取0~1之间,控制过拟合
表3  LightGBM主要参数含义及用法
参数max_depthnum_leaveslearning_ratelambda_l1
取值25650.050.1
表4  LightGBM常用参数取值
图3  研究区不同特征综合的不透水面提取精度对比
原始波段特征1特征1和3特征1和2特征2和3特征1、2和3
阿斯塔纳总体精度OA/%89.9488.1186.8793.0388.8294.00
漏分误差/%5.8614.2017.109.8217.529.78
错分误差/%13.1510.049.954.375.542.40
Kappa系数0.80.760.740.860.780.88
塔什干总体精度OA/%68.6586.7187.2692.0684.2995.75
漏分误差/%55.4113.7314.907.0121.984.80
错分误差/%14.0512.9711.058.7010.793.74
Kappa系数0.370.730.750.840.690.92
杜尚别总体精度OA/%84.5191.0891.4993.6089.4193.98
漏分误差/%24.7513.2013.559.256.808.78
错分误差/%7.645.083.863.7613.363.47
Kappa系数0.690.820.830.870.790.88
表5  研究区不同特征综合的不透水面提取精度
图4  阿斯塔纳不透水面提取结果(图中黑色矩形框与椭圆圈表示不同方法之间差异较明显区域)
图5  不同分类方法不透水面提取结果细节图(图中区域为图4黑色矩形框部分区域)
LightGBMHistGBDTExtratreeRFTreegradXGBoostPISI指数法
总体精度OA/%93.8989.9391.8091.4989.3389.8985.19
漏分误差/%10.0616.849.3012.7617.2115.7417.53
错分误差/%2.343.827.264.654.755.0412.79
Kappa系数0.880.800.840.830.790.800.70
运行时间/s293.831 093.34494.74460.37381.06666.46
表6  不同分类方法不透水面提取精度和运行时间
阿斯塔纳塔什干杜尚别
总体精度OA/%94.0095.7593.98
漏分误差/%9.784.808.78
错分误差/%2.403.743.47
Kappa系数0.880.920.88
表7  研究区城市不透水面提取精度
图6  不透水面产品与本文方法提取结果对比(图中黑色矩形框与椭圆圈表示不同方法之间差异较明显区域)
图7  不透水面产品与本文方法提取结果对比(图中区域为图6黑色矩形框部分区域。阿斯塔纳、塔什干、杜尚别的Google Earth影像拍摄时间分别为2020年9月、2020年8月、2020年8月)
1 ARNOLD C L, GIBBONS C J. Impervious surface coverage: The emergence of a key environmental indicator[J]. Journal of the American Planning Association,1996,62(2):243-258.
2 XU Hanqiu. Quantitative analysis on the relationship of urban impervious surface with other components of the urban ecosystem[J].Acta Ecologica Sinica,2009,29(5): 2456-2462.
2 徐涵秋. 城市不透水面与相关城市生态要素关系的定量分析[J]. 生态学报,2009,29(5):2456-2462.
3 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.
4 WANG Haiyan. Research in comprehensive economic and social development in perspective of “One Belt and One Road”[J]. Journal of Xinjiang Normal University (Philosophy and Social Sciences), 2015,36(5): 78-86.
4 王海燕. “一带一路”视域下中亚国家经济社会发展形势探究[J]. 新疆师范大学学报(哲学社会科学版), 2015, 36(5): 78-86.
5 TAN Zhuting, WANG Xuhong, JIANG Xiaohui, et al. Expansion of megacities in five central Asian countries along “One Belt and One Road” routes (2000-2018)[J]. Journal of Beijing Normal University (Natural Science),2020,56(6): 814-821.
5 谭竹婷, 王旭红, 蒋晓辉,等.2000-2018年“一带一路”之中亚5国首都城市扩张的遥感监测[J]. 北京师范大学学报(自然科学版), 2020, 56(6): 814-821.
6 LI Deren, LUO Hui, SHAO Zhenfeng. Review of impervious surface mapping using remote sensing technology and its aplication[J].Geomatics and Information Science of Wuhan University, 2016, 41(5):569-577,703.
6 李德仁, 罗晖, 邵振峰. 遥感技术在不透水层提取中的应用与展望[J]. 武汉大学学报(信息科学版), 2016, 41(5):569-577,703.
7 RIDD M K. Exploring a V-I-S (vegetation-impervious surface-soil) model for urban ecosystem analysis through remote sensing:Comparative anatomy for cities[J]. International Journal of Remote Sensing, 1995, 16(12):2165-2185.
8 XU Hanqiu. A new remote sensing index for fastly extracting impervious surface information[J]. Geomatics and Information Science of Wuhan University, 2008,33(11): 1150-1153.
8 徐涵秋. 一种快速提取不透水面的新型遥感指数[J]. 武汉大学学报(信息科学版), 2008,33(11):1150-1153.
9 DENG C, WU C. BCI: A biophysical composition index for remote sensing of urban environments[J]. Remote Sensing of Environment, 2012, 127:247-259.
10 TIAN Y, CHEN H, SONG Q, et al. A novel index for impervious surface area mapping: Development and validation[J].Remote Sensing,2018,10(10):1521-1521.DOI:10.3390/rs 10101521
doi: 10.3390/rs 10101521
11 KEBEDE T A, HAILU B T, SURYABHAGAVAN K. V. Evaluation of spectral built-up indices for impervious surface extraction using Sentinel-2A MSI imageries: A case of Addis Ababa city, Ethiopia[J]. Environmental Challenges, 2022, 8: 100568. DOI: 10.1016/j.envc.2022.100568
doi: 10.1016/j.envc.2022.100568
12 SHI L, LING F, GE Y, et al. Impervious Surface change mapping with an uncertainty-based spatial-temporal consistency model: A case study in Wuhan city using Landsat time-series datasets from 1987 to 2016[J]. Remote Sensing, 2017, 9(11):1148. DOI: 10.3390/rs9111148
doi: 10.3390/rs9111148
13 WANG X, ZHOU C, FENG X,et al. Testing the efficiency of using high-resolution data from GF-1 in land cover classifications[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing,2018,11(9):3051-3061.
14 HUANG X, SCHNEIDER A, FRIEDL M A. Mapping sub-pixel urban expansion in China using MODIS and DMSP/OLS nighttime lights[J]. Remote Sensing of Environment, 2016,175:92-108.
15 HU X, WENG Q. Estimating impervious surfaces from medium spatial resolution imagery using the self-organizing map and multi-layer perceptron neural networks[J]. Remote Sensing of Environment, 2009, 113(10): 2089-2102.
16 WANG Jing, GAO Shuai, GUO Liang, et al. Impervious surface extraction from high-resolution images based on multi-scale feature fusion in U-Net network[J]. Remote Sensing Technology and Application,2022,37(4):811-819.
16 王晶,高帅,郭亮,等.基于多尺度特征融合的U-Net网络高分影像不透水面提取研究[J]. 遥感技术与应用,2022,37(4):811-819.
17 ZHANG Hongsheng, LIN Yinyi, WANG Ting, et al. Fusing optical and SAR remote sensing data for urban impervious surface estimation[J]. Geography and Geo-Information Science, 2018, 34(3):39-46.
17 张鸿生,林殷怡,王挺,等.融合光学与雷达遥感数据的城市不透水面提取方法[J]. 地理与地理信息科学, 2018, 34(3):39-46.
18 ZHANG X, LIU L, CHEN X, et al. GLC_FCS30: Global land-cover product with fine classification system at 30 m using time-series Landsat imagery[J]. Earth System Science Data,2021,13(6):2753-2776.
19 GONG P, LI X, WANG J, et al. Annual maps of Global Artificial Impervious Area(GAIA) between 1985 and 2018[J]. Remote Sensing of Environment, 2020, 236(C): 111510. DOI:10.1016/j.rse.2019.111510
doi: 10.1016/j.rse.2019.111510
20 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. DOI: 10.1007/s00521-020-05449-7
doi: 10.1007/s00521-020-05449-7
21 SUN Z, DU W, JIANG H,et al. Global 10-m impervious surface area mapping: A big earth data based extraction and updating approach[J]. International Journal of Applied Earth Observation and Geoinformation,2022,109:102800. DOI:10. 1016/j.jag.2022.102800
doi: 10. 1016/j.jag.2022.102800
22 WU Yonghui, JI Kefeng, YU Wenxian. Comparison of classification performance of full-,dual- and single-polarization SAR images using SVM[J].Journal of Remote Sensing, 2008,12(1):46-53.
22 吴永辉, 计科峰, 郁文贤. 利用SVM的全极化、双极化与单极化SAR图像分类性能的比较[J]. 遥感学报, 2008,12(1):46-53.
23 LEE J-S, GRUNES M R, Pottier E. Quantitative comparison of classification capability: Fully polarimetric versus dual and single-polarization SAR[J]. IEEE Transactions on Geoscience and Remote Sensing, 2001, 39(11):2343-2351.
24 WENG Q. Remote sensing of impervious surfaces in the urban areas: Requirements, methods, and trends[J]. Remote Sensing of Environment,2011,117:34-39. DOI:10.1016/j.rse. 2011.02.030
doi: 10.1016/j.rse. 2011.02.030
25 CHE Meiqin, SAMAT Alim, DU Peijun, et al. Urban man-made target extraction from Quad-PolSAR imagery with rollinvariant parameters[J]. Journal of Remote Sensing,2016,20(2):303-314.
25 车美琴, 阿里木·赛买提, 杜培军, 等. 利用旋转不变特征提取全极化SAR影像人工地物[J]. 遥感学报, 2016,20(2):303-314.
26 LIAO W, MURA M, CHANUSSOT J, et al. Morphological attribute profiles with partial reconstruction[J]. IEEE Transactions on Geoscience and Remote Sensing, 2016, 54(3): 1738-1756.
27 KE G, MENG Q, FINLEY T, et al. LightGBM: a highly efficient gradient boosting decision tree[C]∥ Proceedings of the 31st International Conference on Neural Information Processing Systems. Long Beach, California, USA; Curran Associates Inc. 2017: 3149-3157.
28 LIU Tingyue, DAI Jingjing, ZHAO Yuanyi, et al. Remote sensing inversion of lithium concentration in salt lake using LightGBM:A case study of northern Zabuye salt lake in Tibet[J]. Acta Geologica Sinica,2021,95(7):2249-2256.
28 刘婷玥, 代晶晶, 赵元艺, 等. 基于LightGBM的盐湖锂浓度遥感反演研究——以西藏扎布耶盐湖北湖为例[J]. 地质学报, 2021,95(7):2249-2256.
29 ALIMUJIANG Kamusi, TANG Bing, ANWAER Maimaitiming. Study on the urbanization development characteristics of Central Asia (1960-2009)[J]. Journal of Arid Land Resources and Environment,2013,27(1):21-26.
29 阿里木江·卡斯木, 唐兵, 安瓦尔·买买提明. 近50年来中亚五国城市化发展特征研究[J]. 干旱区资源与环境,2013,27(1):21-26.
30 MA Haitao, SUN Zhan. Comprehensive urbanization level and its dynamic factors of five CentralAsian countries[J]. Acta Geographica Sinica,2021,76(2):367-382.
30 马海涛, 孙湛. 中亚五国综合城镇化水平测度及其动力因素[J]. 地理学报, 2021,76(2):367-382.
31 XIONG Jinhui, YUE Wenze, CHEN Yang, et al. Multi-scenario urban expansion simulation for SDGs: Taking the Central Asian region along the Belt and Road as an example[J]. Journal of Natural Resources,2021,36(4):841-853.
31 熊锦惠, 岳文泽, 陈阳, 等. 面向SDGs的城市扩张多情景模拟——以“一带一路”中亚区为例[J]. 自然资源学报,2021,36(4): 841-853.
32 Vetan TYU. Organic renewal of mahalla community in the Inner city of Tashkent,Uzbekistan[D].Hangzhou: Zhejiang University, 2018.朱越安(TYU VET AN). 乌兹别克斯坦塔什干市旧城马哈拉住区有机更新研究[D]. 杭州: 浙江大学,2018.
33 DALLA-MIRA M, BENEDIKTSSON J A, WASKE B, et al. Morphological attribute profiles for the analysis of very high resolution images[J]. IEEE Transactions on Geoscience and Remote Sensing, 2010, 48(10): 3747-3762.
34 LIAO W, CHANUSSOT J, MURA M D, et al. Taking optimal advantage of fine spatial resolution: Promoting partial image reconstruction for the morphological analysis of very-high-resolution images[J]. IEEE Geoscience and Remote Sensing Magazine, 2017, 5(2):8-28.
[1] 米军,张肖,刘良云. 1985~2020年上海市不透水面时空特征演化分析[J]. 遥感技术与应用, 2023, 38(2): 297-307.
[2] 匡开新,杨英宝,高永年,刘宇翔. 协同Sentinel合成孔径雷达和光学影像多特征的不透水面随机森林提取方法[J]. 遥感技术与应用, 2023, 38(2): 422-431.
[3] 宏格尔,迟文峰,窦银银,郝润梅,包玉海,匡文慧. 近70年内蒙古呼和浩特城市演化的形态特征及影响因素剖析[J]. 遥感技术与应用, 2022, 37(5): 1237-1247.
[4] 王晶,高帅,郭亮,汪云. 基于多尺度特征融合的U-Net网络高分影像不透水面提取研究[J]. 遥感技术与应用, 2022, 37(4): 811-819.
[5] 王姗姗,蒲英霞,李升峰,李闰洁,李茂华. 利用长时序流域不透水面数据集探索城市增长模式——以秦淮河流域为例[J]. 遥感技术与应用, 2022, 37(3): 739-750.
[6] 盖一铭,阿里木·赛买提,王伟,吉力力·阿不都外力. 基于样本迁移的干旱区地表覆盖快速更新[J]. 遥感技术与应用, 2022, 37(2): 333-341.
[7] 李润祥,高小红,汤敏. 基于双树复小波分解的Boosting集成学习土地覆被分类研究[J]. 遥感技术与应用, 2022, 37(2): 354-367.
[8] 符冰雪,张继超,杜文杰,王鹏龙,孙中昶. 一种不透水面精细制图新方法及其在城市SDGs指标评估上的应用[J]. 遥感技术与应用, 2021, 36(6): 1339-1349.
[9] 刘莹,朱秀芳,徐昆,陈令仪,郭锐. 干旱对灌溉和雨养农田生态系统生产力的影响对比分析[J]. 遥感技术与应用, 2021, 36(2): 381-390.
[10] 杨玉婷,汤家法,边金虎,李爱农,雷光斌,黄平,蒋梓淳. 加尔各答市地表温度与不透水面比例季相相关性研究[J]. 遥感技术与应用, 2021, 36(1): 79-89.
[11] 沈谦,朱长明,张新,黄巧华,杨程子. 1992~2013我国干旱区城市不透水遥感制图与扩张过程分析[J]. 遥感技术与应用, 2020, 35(5): 1178-1186.
[12] 刘美,杜国明,于凤荣,匡文慧. 哈尔滨城乡梯度建设用地结构变化及不透水面遥感监测分析[J]. 遥感技术与应用, 2020, 35(5): 1206-1217.
[13] 刘婵,刘冰,赵文智,朱钊岑,司瑞. 中亚地区植被净初级生产力时空动态及其与气候因子关系[J]. 遥感技术与应用, 2020, 35(4): 924-933.
[14] 李智礼,匡文慧,张澍. 近70 a天津主城区城市土地利用/覆盖变化遥感监测与时空分析[J]. 遥感技术与应用, 2020, 35(3): 527-536.
[15] 陈馨,匡文慧. 基于云平台的中哈干旱区典型城市地表覆盖变化遥感监测与比较[J]. 遥感技术与应用, 2020, 35(3): 548-557.