基于MODIS影像的土地覆被分类研究——以京津冀地区为例

doi:10.11820/dlkxjz.2014.11.012

摘要/Abstract

摘要： 在全球变化研究中,如何快速、准确获取土地覆被信息对该项研究有着至关重要的作用.随着遥感科学的不断发展和应用领域的深入,研究者可以利用遥感影像进行土地覆被分类研究,并且具有准确、快速、自动化等优点.本文利用MODIS数据具有的多光谱、多时相特点,以京津冀地区为例,选取2013 年全年16-day 的MOD13Q1/EVI时间序列数据、2013 年5 月份一期的MOD09Q1(1、2 波段)和MOD09A1(3-7 波段)产品,并运用时间序列谐波分析法对全年MOD13Q1/EVI 时间序列数据进行去云、去噪的平滑重建处理,使其数据更能反映物候周期性变化规律.选择谐波分析后的全年MOD13Q1/EVI 时间序列数据、MODIS数据的1-7 波段地表反射率和NDWI(归一化差异水体指数)、MNDWI(改进归一化差异水体指数)和NDSI(土壤亮度指数),构建了3 种特征变量组合方案的CART决策树,分别进行京津冀地区的土地覆被分类研究.结果表明:方案一(全年EVI 的23 个时相)、方案二(方案一+MOD09 的1-7 波段地表反射率)和方案三(方案二+MNDWI+NDSI+NDWI)的总体分类精度分别达到86.70%、89.98%、91.34%,Kappa系数分别为84.94%、88.66%、90.20%.研究表明,仅利用MODIS遥感影像自身多种分类特征和决策树方法对宏观土地覆被分类就可达到较高精度,显示了本文分类方法在实践中的可行性及MODIS数据在区域尺度土地覆被分类研究方面的优势与潜力.

关键词: CART决策树, MODIS影像, 分类, 分类特征组合, 京津冀地区, 土地覆被, 谐波分析

Abstract: With intensifying human activities global ecological and environmental problems have become increasingly pressing. Therefore the study of global change has become more prominent. Obtaining accurate land cover information in a timely fashion is critically important for such research. With the development of remote sensing science and technology and application, numerous studies have the issue of land cover classification using remote sensing image. In this study, the Beijing-Tianjin-Hebei region was selected as a study area for land cover classification using MODIS data. The 16-day MOD13Q1/EVI data in 2013, MOD09Q1 (Band1, 2) and MOD09A1(Band3- 7) products in May 2013 were used as the basic data. Harmonic analysis method was employed to remove clouds and noises of the whole year's EVI, so that it can better reflect plant phenology. Then, the processed MOD13Q1/EVI data, surface albedo of MODIS/Ref1-7, modified normalized difference water index (MNDWI), normalized difference soil brightness index (NDSI), and normalized difference water index (NDWI) were integrated to construct three schemes of CART decision tree to investigate the land cover classification of the Beijing-Tianjin-Hebei region. The three band combinations are scheme1: 23 phases of EVI of 2013; scheme2: 23 phases of EVI of 2013 plus B1-B7 of MOD09; and scheme3: scheme2+MNDWI+NDSI+NDWI. The results show that the overall classification accuracy of the three schemes reaches 86.70%, 89.98%, and 89.98% respectively. Their Kappa coefficients are 84.94%, 88.66%, and 90.20% respectively. Therefore, MODIS images with various classification characteristics, combined with the decision trees can achieve higher precision land cover classification. This proves the feasibility of the proposed method for land cover classification at the regional scale.

Key words: Beijing-Tianjin-Hebei region, CART decision tree, classification, classification feature combination, harmonics analysis, land cover, MODIS images

中图分类号:

TP79

左玉珊, 王卫, 郝彦莉, 刘红. 基于MODIS影像的土地覆被分类研究——以京津冀地区为例[J]. 地理科学进展, 2014, 33(11): 1556-1565.

ZUO Yushan, WANG Wei, HAO Yanli, LIU Hong. Land cover classification based on MODIS images: taking the Beijing-Tianjin-Hebei region as an example[J]. PROGRESS IN GEOGRAPHY, 2014, 33(11): 1556-1565.

参考文献

陈云, 戴锦芳, 李俊杰. 2008. 基于影像多特征的CART 决策树分类方法及其应用. 地理与地理信息科学, 24(2): 33-36. [Chen Y, Dai J F, Li J J. 2008. CART-based decisiontree classifier using multi-feature of image and its application.Geography and Geo-Information Science, 24(2): 33-36.]
金卫斌, 熊勤学, 薛莲. 2011. 基于MODIS-EVI时序数据的江汉平原四湖地区土地覆被动态分析. 湖北农业科学, 50(11): 2220-2224. [Jin W B, Xiong Q X, Xue L. 2011. Dynamicanalysis of the land cover in Four-Lake area in JianghanPlain based on MODIS-EVI time-series data. HubeiAgricultural Sciences, 50(11): 2220-2224.]
孔凡明, 蒋卫国, 李京, 等. 2013. 基于MODIS的2011 年泰国洪涝受灾信息提取与分析. 灾害学, 28(2): 95-99. [KongF M, Jiang W G, Li J, et al. 2013. Extraction and analysisof Thailand flood affected region in 2011 based on MODISdata. Journal of Catastrophology, 28(2): 95-99.]
李红军, 郑力, 雷玉平. 2007. 基于EOS/MODIS数据的NDVI与EVI 比较研究. 地理科学进展, 26(1): 26-32. [Li H J,Zheng L, Lei Y P. 2007. Comparison of NDVI and EVIbased on EOS/MODIS data. Progress Progress in Geography,26(1): 26-32.]
李淼. 2007. 基于支持向量机的MODIS 数据土地覆被分类研究[D]. 阜新: 辽宁工程大学. [Li M. 2007. Land coverclassification with SVM applied to MODIS imagery[D].Fuxin, China: Liaoning Technical University.]
李文梅, 覃志豪, 杨强. 2010. MODIS NDVI 与MODIS EVI的比较分析. 遥感信息, (6): 73-78. [Li W M, Tan Z H,Yang Q. 2010. Comparison and analysis of MODIS NDVIand MODIS EVI. Remote Sensing Information, (6):73-78.]
刘爱霞, 王静, 吕春艳. 2006. 基于MODIS 数据的北京西北部地区土地覆被分类研究. 地理科学进展, 25(2): 96-102. [Liu A X, Wang J, Lv C Y. 2006. Land cover classificationbased on MODIS data in area to the north-west ofBeijing. Progress in Geography, 25(2): 96-102.]
刘建光, 李红, 孙丹峰, 等. 2010. MODIS土地利用/土地覆被多时相多光谱决策树分类. 农业工程学报, 26(10): 312-318. [Liu J G, Li H, Sun D F, et al. 2010. Land use/coverdecision tree classification fusing multi- temporal andmulti- spectral of MODIS. Transactions of the CSAE, 26(10): 312-318.]
刘纪远. 1996. 中国资源遥感宏观调查与动态研究. 北京: 中国科学技术出版社. [LIU J Y. 1996. Zhongguo ziyuanyaogan hongguan diaocha yu dongtai yanjiu. Beijing, China:China Science and Technology Press .]
刘勇洪, 牛铮. 2004. 基于MODIS遥感数据的宏观土地覆被特征分类方法与精度分析研究. 遥感技术与应用, 19(4): 217-224. [Liu Y H, Niu Z. 2004. Regional land coverimage classification and accuracy evaluation using MODISdata. Remote Sensing Technology and Application,19(4): 217-224.]
刘勇洪, 牛铮, 徐永明, 等. 2006. 基于MODIS数据设计的中国土地覆被分类系统与应用研究. 农业工程学报, 22(5): 99-105. [Liu Y H, Niu Z, Xu Y M, et al. 2006. Designof land cover classification system for China and itsapplication research based on MODIS data. Transactionsof the CSAE, 22(5): 99-105.]
马娜, 胡云峰, 庄大方, 等. 2010. 基于最佳波段指数和JM距离可分性的高光谱数据最佳波段组合选取研究. 遥感技术与应用, 25(3): 358-365. [Ma N, Hu Y F, Zhuang DF, et al. 2010. Determination on the optimum band combinationof HJ-1A hyperspectral data in the case region ofDongguan based on optimum index factor and J-M distance.Remote Sensing Technology and Application, 25(3): 358-365.]
马少平, 朱小燕. 2004. 人工智能. 北京: 清华大学出版社.[Ma S P, Zhu X Y. 2004. Rengong zhineng. Beijing, China:Tsinghua University Press.]
那晓东, 张树清, 李晓峰, 等. 2007. MODIS NDVI 时间序列在三江平原湿地植被信息提取中的应用. 湿地科学, 5(3): 227-236. [Na X D, Zhang S Q, Li X F, et al. 2007.Application of MODS NDVI time series to extractingwetland vegetation information in the Sanjiang Plain.Wetland Science, 5(3): 227-236.]
申文明, 王文杰, 罗江海, 等. 2007. 基于决策树分类技术的遥感影像分类方法研究. 遥感技术与应用, 22(3): 333-338. [Shen W M, Wang W J, Luo J H, et al. 2007. Classificationmethods of remote sensing image based on decisiontree technologies. Remote Sensing Technology andApplication, 22(3): 333-338.]
王大鹏, 王周龙, 李德一, 等. 2007. 综合非光谱信息的荒漠化土地CART 分类. 遥感学报, 11(4): 487-492. [Wang DP,Wang Z L, Li D Y, et al. 2007. The extraction of desertificationinformation using CART under knowledge guide.Journal of Remote Sensing, 11(4): 487-492.]
魏强. 2009. 基于MODIS和TM数据的京津冀地区土地覆被分类方法研究[D]. 石家庄: 河北师范大学[Wei Q. 2009.Research on land cover classification method of Beijing-Tianjin-Heibei region using MODIS and TM data[D]. Shijiazhuang,China: Hebei Normal University.]
徐涵秋. 2005. 利用改进的归一化差异水体指数(MNDWI)提取水体信息的研究. 遥感学报, 9(5): 589-595. [Xu H Q.2005. A study on information extraction of water bodywith the modified normalized difference water index(MNDWI). Journal of Remote Sensing, 9(5): 589-595.]
延昊. 2002. 中国土地覆被变化与环境影响遥感研究[D]. 北京: 中国科学院遥感应用研究所. [Yan H. 2002. Remotesensing study of land cover change and its environmentalin China[D]. Beijing, China: Institute of Remote SensingApplications, CAS.]
岳瑞红. 2010. 基于MODIS 数据的蒙古高原土地覆被分类研究[D]. 呼和浩特: 内蒙古师范大学. [Yue R H. 2010.Research on land cover classification in Mongolian Plateaubased on MODIS data[D]. Hohhot, China: InnerMongolia Normal University.]
张会, 闫金凤. 2013. 基于MODIS 影像多特征的CART 决策树分类. 地理空间信息, 11(2): 111-113. [Zhang H, Yan JF. 2013. CART decision tree classifier based on multi-featureof MODIS data. Geospatial Information, 11(2): 111-113.]
张楠楠, 王文, 王胤. 2012. 鄱阳湖面积的卫星遥感估计及其与水位关系分析. 遥感技术与应用, 27(6): 947- 953.[Zhang N N, Wang W, Wang Y. 2012. Estimate the areaof the Poyang Lake using statellite remote sensing dataand analyze its relationship with water level. RemoteSensing Technology and Application, 27(6): 947-953.]
赵萍, 傅云飞, 郑刘根, 等. 2005. 基于分类回归树分析的遥感影像土地利用/覆被分类研究. 遥感学报, 9(6): 708-716. [Zhao P, Fu Y F, Zheng L G, et al. 2005. Cart-basedland use/cover classification of remote sensing images.Journal of Remote Sensing, 9(6): 708-716.]
赵英时. 2001. 美国中西部沙山地区环境变化的遥感研究.地理研究, 20(2): 213-219. [Zhao Y S. 2001. A study onenvironmental change analysis in Sand Hill of Nebraskausing remote sensing. Geographical Research, 20(2): 213-219.]
赵英时. 2003. 遥感应用分析原理与方法. 北京: 科学出版社. [Zhao Y S. 2003. Yaogan yingyong fenxi yuanli yufangfa. Beijing, China: Science Press.]
Chen Y, Huang C, Ticehurst C, et al. 2013. An evaluation of MODIs daily and 8- day composite products for floodplain and wetland inundation mapping. Wetlands, 33(5): 823-835.
DeFries R S, Hansen M, Towsend J G R, et al. 1998. Global land coverclassification at 8 km spatial resolution: the use of training data from landsat imagery in decision tree classifies. International Journal of Remote Sensing, 19(6): 3141-3168.
Friedl MA, McIver D K, Hodges J C F. 2002. Global land cover mapping from MODIS: algorithms and early results. Remote Sensing of Environment, 83: 287-302.
Gopal S, Woodcock C E, Strahler A H. 1999. Fuzzy neural network classification of global land cover from a 1° AVHRR data set. Remote Sensing of Environment, 67(2): 230-243.
Julien Y, Sobrino J A, Verhoef W. 2006. Changes in land surface temperatures and NDVI values over Europe between 1982 and 1999. Remote Sensing of Environment, 103(1): 43-55.
Mackin K J, Nunohiro E, Ohshiro M, et al. 2006. Land cover classification from MODIS satellite data using probabilisti- cally optimal ensemble of artificial neural networks. Lecture Notes in Computer Science, 4253: 820-826.
Muchoney D, Borak J. 2000. Application of the MODIS global supervised classification refel to vegetation and land cover mapping of central America. International Journal of Remote Sensing, 21: 1115-1138.
Price J C. 2003. Compare MODIS and ETM+ data for regional and global land classification. Remote Sensing of Environment, 86(4): 491-499.
Roerink G J, Menen M, Verhoef W. 2000. Reconstructing cloudfree NDVI composites using Fourier analysis of time series. International Journal of Remote Sensing, 21 (9): 1911-1917.
Vogelmann J E, Sohl T L, Campbell P V. 1998. Regional characterization of land cover using multiple sources of data. Photogrammetric Engineering and Remote Sensing, 64 (1): 45-57.

[1]	张华, 刘哲达, 殷小冰. 中国跨省流动人口回流意愿的空间差异及影响因素[J]. 地理科学进展, 2021, 40(1): 73-84.
[2]	刘采, 张海燕, 李迁. 1980—2018年海南岛人类活动强度时空变化特征及其驱动机制[J]. 地理科学进展, 2020, 39(4): 567-576.
[3]	唐寅, 王中根, 王婉清, 黄火键, 袁勇. 适用于遥感影像的水生态空间多功能分类体系研究[J]. 地理科学进展, 2020, 39(3): 454-460.
[4]	付慧, 刘艳军, 孙宏日, 周国磊. 京津冀地区耕地利用转型时空分异及驱动机制[J]. 地理科学进展, 2020, 39(12): 1985-1998.
[5]	郭新伟, 余斌, 卓蓉蓉, 曾菊新, 颜梅艳. 江汉平原乡村农户就业变迁及其影响因素研究[J]. 地理科学进展, 2020, 39(12): 2094-2104.
[6]	黄琳珊, 陈彦光, 李双成. 京津冀城镇用地空间结构的多分维谱分析[J]. 地理科学进展, 2019, 38(1): 50-64.
[7]	钟炜菁, 王德. 基于居民行为周期特征的城市空间研究[J]. 地理科学进展, 2018, 37(8): 1106-1118.
[8]	李宵, 申玉铭, 邱灵. 京津冀生产性服务业关联特征分析[J]. 地理科学进展, 2018, 37(2): 299-307.
[9]	张雅春, 那晓东, 臧淑英. 基于HJ-1A高光谱影像的湿地精细分类[J]. 地理科学进展, 2018, 37(12): 1705-1712.
[10]	左秀玲, 苏奋振, 赵焕庭, 方月, 杨娟. 南海珊瑚礁高分辨率遥感地貌分类体系研究[J]. 地理科学进展, 2018, 37(11): 1463-1472.
[11]	张甘霖, 朱阿兴, 史舟, 王秋兵, 刘宝元, 张兴昌, 史志华, 杨金玲, 刘峰, 宋效东, 吴华勇, 曾荣. 土壤地理学的进展与展望[J]. 地理科学进展, 2018, 37(1): 57-65.
[12]	徐勇, 张雪飞, 周侃, 王传胜, 徐小任. 资源环境承载能力预警的超载成因分析方法及应用[J]. 地理科学进展, 2017, 36(3): 277-285.
[13]	刘年磊, 卢亚灵, 蒋洪强, 程曦. 基于环境质量标准的环境承载力评价方法及其应用[J]. 地理科学进展, 2017, 36(3): 296-305.
[14]	徐卫华, 杨琰瑛, 张路, 肖燚, 王效科, 欧阳志云. 区域生态承载力预警评估方法及案例研究[J]. 地理科学进展, 2017, 36(3): 306-312.
[15]	孟晖, 李春燕, 张若琳, 李亚民. 京津冀地区县域单元地质灾害风险评估[J]. 地理科学进展, 2017, 36(3): 327-334.