Please wait a minute...
img

官方微信

遥感技术与应用  2021, Vol. 36 Issue (4): 936-947    DOI: 10.11873/j.issn.1004-0323.2021.4.0936
遥感应用     
基于Sentinel-2的UNVI植被指数及性能对比研究
朱曼1,2(),张立福1(),王楠1,林昱坤1,2,张琳姗1,2,王飒1,2,刘华亮3
1.中国科学院空天信息创新研究院,北京 100094
2.中国科学院大学,北京 100049
3.北京大学遥感与地理信息系统研究所,北京 100871
Comparative Study on UNVI Vegetation Index and Performance based on Sentinel-2
Man Zhu1,2(),Lifu Zhang1(),Nan Wan1,Yukun Lin1,2,Linshan Zhang1,2,Sa Wang1,2,Hualiang Liu3
1.Arerospace Information Research Institute,Chinese Academy of Sciences,Beijing 100094,China
2.University of Chinese Academy of Sciences,Beijing 100049,China
3.Institute of Remote Sensing and Geographic Information Systems,Peking University,Beijing 100871,China
 全文: PDF(4906 KB)   HTML
摘要:

作物精准识别和分类是农业遥感检测的重要内容,对作物长势监测以及估产十分重要。以美国混合农业带为研究区,基于Sentinel-2时间序列影像,根据其传感器响应函数计算了针对Sentinel-2的通用归一化植被指数(Universal Normalized Vegetation Index,UNVI),并通过两个对比实验,分析UNVI等6个指数在作物精准分类中的性能。实验一以JM(Jeffries-Matusita)距离为指标对不同作物类别之间的可分性进行分析,结果表明UNVI优于NDVI、EVI、WDRVI、NDre1和NDWI指数,在玉米和棉花、玉米和水稻、玉米和水稻的区分上,UNVI优于其他指数区分能力相当,但在其余的作物组合上如棉花和水稻,NDVI等指数则无法将其很好的区分,此时UNVI指数依然可以表现出较好的区分能力;实验二对6种时间序列指数特征分别使用随机森林和支持向量机进行作物分类,结果表明UNVI指数的总体精度和Kappa系数最高,其次是NDre1指数和WDRVI指数,EVI的总体精度和Kappa系数最低,这表明UNVI比其他6个指数更好地区分了研究区大豆、玉米、棉花和水稻等4种主要作物。综上,基于Sentinel-2时间序列的UNVI指数在进行作物分类时与其他5种遥感植被指数相比,具有较大的优势,UNVI可为农作物长势分析和作物估产研究等农业研究和应用的可选植被指数。

关键词: Sentinel?2时间序列UNVI植被指数可分性作物识别    
Abstract:

Accurate crop identification and classification is an important part of agricultural remote sensing detection, which is very important for crop growth monitoring and yield estimation. In this paper, based on the Sentinel-2 time series images of the United States mixed agricultural belt as the research area, the Universal Normalized Vegetation Index (UNVI) for Sentinel-2 is calculated according to its sensor response function, and two comparisons are made. Experiment to analyze the performance of UNVI and other six indexes in the accurate classification of crops. Experiment 1 uses the JM (Jeffries-Matusita) distance as an indicator to analyze the separability between different crop categories. The results show that UNVI is better than NDVI, EVI, WDRVI, NDre1 and NDWI index. In corn and cotton, corn and rice, In terms of distinguishing between corn and rice, UNVI is better than other indexes in distinguishing ability, but in other crop combinations such as cotton and rice, NDVI and other indexes cannot distinguish them well. At this time, UNVI index can still perform better Distinguishing ability of experiment; Experiment 6 uses random forests and support vector machines to classify crops of the six time series index features. The results show that the UNVI index has the highest overall accuracy and Kappa coefficient, followed by the NDre1 index and the WDRVI index, and the EVI overall accuracy and The Kappa coefficient is the lowest, which indicates that UNVI distinguishes the four main crops of soybean, corn, cotton and rice in the study area better than the other five indexes. In summary, the UNVI index based on the Sentinel-2 time series has greater advantages in crop classification than other remote sensing vegetation indexes studied in this paper. UNVI can be used for agricultural research and application such as crop growth analysis and crop yield research Optional vegetation index.

Key words: Sentinel-2    Time series    UNVI vegetation index    Separability    Crop identification
收稿日期: 2020-06-12 出版日期: 2021-09-26
ZTFLH:  TP79  
基金资助: 国家自然科学基金重点基金项目(41830108);兵团重点领域创新团队建设计划(2018CB004);兵团重大科技课题(2018A A00402);中国科学院战略性先导科技专项(XDA19080304)
通讯作者: 张立福     E-mail: zhuman@radi.ac.cn;zhanglf@radi.ac.cn
作者简介: 朱曼(1994-),女,安徽宿州人,硕士研究生,主要从事高光谱遥感研究。E?mail:zhuman@radi.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  
朱曼
张立福
王楠
林昱坤
张琳姗
王飒
刘华亮

引用本文:

朱曼,张立福,王楠,林昱坤,张琳姗,王飒,刘华亮. 基于Sentinel-2的UNVI植被指数及性能对比研究[J]. 遥感技术与应用, 2021, 36(4): 936-947.

Man Zhu,Lifu Zhang,Nan Wan,Yukun Lin,Linshan Zhang,Sa Wang,Hualiang Liu. Comparative Study on UNVI Vegetation Index and Performance based on Sentinel-2. Remote Sensing Technology and Application, 2021, 36(4): 936-947.

链接本文:

http://www.rsta.ac.cn/CN/10.11873/j.issn.1004-0323.2021.4.0936        http://www.rsta.ac.cn/CN/Y2021/V36/I4/936

图1  研究区位置
获取日期获取卫星影像质量
2018-04-21Sentinel-2B较好
2018-05-01Sentinel-2B较好
2018-05-11Sentinel-2B较好
2018-05-16Sentinel-2A良好
2018-06-05Sentinel-2A较好
2018-06-10Sentinel-2B较好
2018-06-15Sentinel-2A良好
2018-06-30Sentinel-2B良好
2018-07-05Sentinel-2A良好
2018-07-10Sentinel-2B良好
2018-07-20Sentinel-2B较好
2018-07-25Sentinel-2A良好
2018-08-04Sentinel-2A较好
2018-09-03Sentinel-2A良好
2018-09-13Sentinel-2A较好
2018-09-18Sentinel-2B良好
2018-09-28Sentinel-2B良好
2018-10-03Sentinel-2A良好
2018-10-18Sentinel-2B较好
2018-10-23Sentinel-2A较好
表1  S2A和S2B获取时间
波段Sentinel-2ASentinel-2B

分辨率

/m

中心波长

/nm

波段宽度

/nm

中心波长

/nm

波段宽度

/nm

1 海岸波段443.927442.34560
2 蓝波段496.698492.19810
3 绿波段560.0455594610
4 红波段664.5386653910
5 植被红边1波段703.919703.82020
6 植被红边2波段740.218739.11820
7 植被红边3波段782.528779.72820
8 近红外波段(宽)835.114583313310
8a 近红外波段(窄)864.8338643220
9 水汽波段945.026943.22760
10 卷云波段1 373.5751 376.97660
11 短波红外11 613.71431 610.414120
12 短波红外22 202.42422 185.723820
表2  S2A和S2B详细信息对比
指数计算公式
通用归一化植被指数UNVIUNVI=Cv-a?Cs-C4Cw+Cv+Cs
归一化植被指数NDVINDVI=ρNIR-ρREDρNIR+ρRED
增强型植被指数EVIEVI=2.5×ρNIR-ρREDρNIR+6×ρRED-7.5×ρBLUE+1
宽动态植被指数WDRVIWDRVI=0.2×ρNIR-ρRED0.2×ρNIR+ρRED
归一化红边指数NDre1NDre1=ρRED2-ρRED1ρRED2+ρRED2
归一化水体指数NDWINDWI=ρGREEN-ρNIRρGREEN+ρNIR
表3  用于对比分析的指数计算公式
图2  技术路线图
图3  4种作物的不同指数时间序列曲线
图4  单时相下每个类对之间的JM距离
图5  多时相下每个类对之间的JM距离
图6  不同指数下4种作物分类的总体精度(OA)和Kappa系数比较
UNVINDVIEVIWDRVINDre1NDWI
PAUAPAUAPAUAPAUAPAUAPAUA
玉米89.894.0388.294.2387.892.9188.694.368994.3888.194.63
棉花91.385.3386.783.9385.784.68985.498985.2987.385.93
水稻92.297.4687.296.6785.196.788.297.678897.1386.397.4
大豆94.165.1793.159.4192.756.9893.960.7493.361.8394.958.36
表4  不同指数下的每种作物分类的制图精度和用户精度 (%)
图7  不同指数下四种作物分类的总体精度(OA)和Kappa系数比较
UNVINDVIEVIWDRVINDre1NDWI
PAUAPAUAPAUAPAUAPAUAPAUA
玉米80.289.9180.588.7579.285.8181.389.548086.6781.384.07
棉花82.670.4271.564.3660.27077.168.1178.164.0774.172.01
水稻85.196.769.592.9145.394.1876.294.076890.9167.593.75
大豆87.352.3184.745.1385.535.3786.347.683.747.4285.944.19
表5  不同指数下的每种作物分类的制图精度和用户精度 (%)
1 Begue A, Arvor D, Bellon B, et al. Remote sensing and cropping practices: a review[J]. Remote Sensing,2018,10:99. DOI:10.3390/rs10010099.
doi: 10.3390/rs10010099
2 Jia K, Liang S, Wei X, et al. Land cover classification of landsat data with phenological features extracted from time series MODIS NDVI data [J]. Remote Sensing, 2014, 6(11): 11518-11532. DOI:10.3390/rs61111518.
doi: 10.3390/rs61111518
3 Knight J F, Lunetta R S, Ediriwickrema J, et al. Regional scale land cover characterization using MODIS-NDVI 250 m Multi-Temporal imagery: a phenology-based approach[J]. Mapping Sciences & Remote Sensing, 2006, 43(1): 1-23. DOI:10.2747/1548-1603.43.1.1
doi: 10.2747/1548-1603.43.1.1
4 Kong F, Li X, Hong W, et al. Land Cover classification based on fused data from GF-1 and MODIS NDVI time series[J]. Remote Sensing,2016,8(9):741. DOI: 10.3390/rs8090741.
doi: 10.3390/rs8090741
5 Liu K, Su H, Zhang L, et al. Analysis of the urban heat island effect in Shijiazhuang, China using satellite and airborne data[J]. Remote Sensing, 2015, 7(4): 4804-4833. DOI:https:10.3390/rs70404804.
doi: https:10.3390/rs70404804
6 Yang Shao, Ross S, Brandon Lunetta , et al. An evaluation of time-series smoothing algorithms for land-cover classifications using MODIS-NDVI multi-temporal data[J]. Remote Sensing of Environment, 2016, 174: 258-265. DOI:10.1016/j.rse.2015.12.023.
doi: 10.1016/j.rse.2015.12.023
7 Brown J C, Kastens J H, Coutinho A C, et al. Classifying Multiyear Agricultural land use data from Mato Grosso using time-series MODIS vegetation index data[J]. Remote Sensing of Environment, 2013, 130: 39-50. DOI:10.1016/j.rse.2012.11.009.
doi: 10.1016/j.rse.2012.11.009
8 Arovor D, Jonathan M, Meirelles M S P, et al. Classification of MODIS EVI time series for crop mapping in the state of Mato Grosso, Brazil[J]. International Journal of Remote Sensing,2011,32(22):7847-7871. DOI:10.1080/01431161. 2010.531783.
doi: 10.1080/01431161. 2010.531783
9 Bendini H N, Fonseca L M G, Schwieder M, et al. Detailed agricultural land classification in the brazilian cerrado based on phenological information from dense satellite image time series [J]. International Journal of Applied Earth Observation and Geoinformation,2019,82:101872. DOI:10.1016/j.jag.2019. 05.005.
doi: 10.1016/j.jag.2019. 05.005
10 Clark M L, Aide T M , Grau H R, et al. A Scalable approach to mapping annual land cover at 250 m using MODIS time series data: a case study in the dry chaco ecoregion of south America[J]. Remote Sensing of Environment, 2010, 114(11): 2816-2832. DOI:10.1016/j.rse.2010.07.001.
doi: 10.1016/j.rse.2010.07.001
11 Peng G, Jie W, Le Y, et al. Finer resolution observation and monitoring of global land cover: first mapping results with Landsat TM and ETM+ data[J]. International Journal for Remote Sensing, 2013, 34(7-8):2607-2654.
12 Senf C, Pflugmacher D, Linden Der, S V, et al. Mapping rubber plantations and natural forests in Sishuangbanna (southwest China) using multi-spectral phenological metrics from MODIS time series[J]. Remote Sensing, 2013, 5: 2795-2812. DOI:10.3390/rs5062795.
doi: 10.3390/rs5062795
13 Wolter P T, Mladenoff D J, Host G E, et al. Improved forest classification in the Northern Lake States using Multi-Temporal Landsat imagery [J]. Photogrammetric Engineering and Remote Sensing, 1995, 61(9): 1129-1143.
14 Qiong H U, Wenbin W U, Song Q, et al. How do temporal and spectral features matter in crop classification in Heilongjiang province,China[J]. Journal of Integrative Agriculture, 2017, 16(2): 324-336. DOI:10.1016/S2095-3119(15)61321-1.
doi: 10.1016/S2095-3119(15)61321-1
15 Zhang L, Furumi S, Muramatsu K, et al. A new vegetation index based on the universal pattern decomposition method [J]. International Journal of Remote Sensing, 2007,28(1-2): 107-124.DOI:10.1080/01431160600857402.
doi: 10.1080/01431160600857402
16 Jiao W, Zhang L, Chang Q, et al. Evaluating an enhanced Vegetation Condition Index (VCI) based on VIUPD for drought monitoring in the Continental United States[J]. Remote Sensing, 2016, 8(3): 224. DOI:10.3390/rs8030224.
doi: 10.3390/rs8030224
17 Su Wei, Zhang Mingzheng, Jiang Kunping, et al. Atmospheric dorrection method for Sentinel-2 satellite imagery[J]. Acta Agronomica Sinica, 2018, 38(1): 0128001.苏伟, 张明政, 蒋坤萍, 等. Sentinel-2卫星影像的大气校正方法[J]. 光学学报, 2018, 38(1): 0128001. DOI:10.1117/3.2533493.
doi: 10.1117/3.2533493
18 Han W, Yang Z, Di L, et al. Crop Scape: A web service based Application for exploring and disseminating US conterminous geospatial dropland data products for secision aupport[J].Computers & Electronics in Agriculture,2012,84:111-123.DOI:10.1016/j.compag.2012.03.005.
doi: 10.1016/j.compag.2012.03.005
19 Hansen M C, Loveland, T R. A review of large area monitoring of land cover change using Landsat data[J]. Remote Sensing of Environment, 2012, 122: 66-74. DOI:10.1016/j.rse.2011.08.024.
doi: 10.1016/j.rse.2011.08.024
20 Boryan C, Yang Z, Mueller R, et al. Monitoring US agriculture: the US department of agriculture, national agricultural statistics service, cropland data layer program[J]. Geocarto International,2011,26(5):341-358. DOI:10.1080/10106049. 2011.562309.
doi: 10.1080/10106049. 2011.562309
21 Zhang L, Qiao N, Baig M H A, et al. Monitoring vegetation dynamics using the Universal Normalized Vegetation Index (UNVI): an optimized vegetation index-VIUPD[J]. Remote Sensing Letters, 2019, 10(7-9): 629-638. DOI:10.1080/2150704X.2019.1597298.
doi: 10.1080/2150704X.2019.1597298
22 Huete A. A comparison of vegetation indices over a global set of TM images for EOS-MODIS[J]. Remote Sensing of Environment, 1997, 59(3): 440-451. DOI:10.1016/S0034-4257(96)00112-5.
doi: 10.1016/S0034-4257(96)00112-5
23 Fernandez-Manso, Oscar, Quintano, et al. Sentinel-2A Red-Edge spectral indices suitability for discriminating burn severity[J]. International Journal of Applied Earth Observation & Geoinformation,2016,50:170-175. DOI:/10.1016/j.jag.2016. 03.005.
doi: /10.1016/j.jag.2016. 03.005
24 Liu Shu, Jiang Qigang, Ma Yue, et al. Object-oriented setland classification based on hybrid feature selection method combining with relief f, multi-objective genetic algorithm and random forest[J]. Transactions of the Chinese Society for Agricultural Machinery, 2017, 48(1): 119-127.
24 刘舒, 姜琦刚, 马玥, 等. 基于多目标遗传随机森林特征选择的面向对象湿地分类 [J]. 农业机械学报, 2017, 48(1): 119-127.
25 Hofmann W. Remote sensing: the quantitative approach[J]. Earth Science Reviews, 1980, 16:0-387. DOI:10.1016/0012-8252(80)90089-6.
doi: 10.1016/0012-8252(80)90089-6
26 Yeom J, Han Y, Kim Y. Separability analysis and classification of rice fields using KOMPSAT-2 High resolution satellite imagery[J]. Research Journal of Chemistry & Environment, 2013, 17(12): 136-144.
27 Deng C, Wu C. BCI: A Biophysical composition index for remote sensing of urban environments[J]. Remote Sensing of Environment,2012,127:247-259. DOI:10.1016/j.rse.2012. 09.009.
doi: 10.1016/j.rse.2012. 09.009
28 Hu Q, Sullamenashe D, Xu B. et al. A Phenology-based spectral and temporal feature selection Mmthod for crop mapping from satellite time series[J]. International Journal of Applied Earth Observation and Geoinformation, 2019, 80: 218-229. DOI:10.1016/j.jag.2019.04.014.
doi: 10.1016/j.jag.2019.04.014
29 Kaufman Y J, Remer L A. Detection of forests using Mid-IR Reflectance: an application for aerosol studies[J]. IEEE Transactions on Geoscience and Remote Sensing, 1994, 32(3): 672-683. DOI: 10.1109/36.297984.
doi: 10.1109/36.297984
30 Somers B, Asner G P. Multi-Temporal hyperspectral mixture analysis and feature selection for invasive species mapping in rainforests[J]. Remote Sensing of Environment, 2013,136:14-27. DOI:10.1016/j.rse.2013. 04.006.
doi: 10.1016/j.rse.2013. 04.006
31 Chen L, Jin Z, Michishita R, et al. Dynamic monitoring of wetland cover changes using time-series remote sensing imagery[J]. Ecological Informatics, 2014, 24: 17-26. DOI:10.1016/j.ecoinf.2014.06.007.
doi: 10.1016/j.ecoinf.2014.06.007
32 Thomas I L, Ching N P, Benning V M, et al. Review article a review of multi-channel indices of class separability[J]. International Journal of Remote Sensing, 1987, 8(3): 331-350. DOI:10.1080/01431168708948645.
doi: 10.1080/01431168708948645
33 Auffarth B, López M, Cerquides J. Comparison of redundancy and relevance measures for feature selection in tissue classification of CT images[C]∥ Proceedings of the 10th Industrial Conference on Advances in Data Mining: Applications and Theoretical Aspects, 1970. DOI:10.1007/978-3-642-14400-4_20.
doi: 10.1007/978-3-642-14400-4_20
34 Belgiu M, Dragut L. Random forest in remote sensing: a review of applications and future directions[J]. ISPRS Journal of Photogrammetry & Remote Sensing, 2016, 114(4): 24-31. DOI:10.1016/j.isprsjprs.2016.01.011.
doi: 10.1016/j.isprsjprs.2016.01.011
35 Nitze I, Barrett B, Cawkwell F. Temporal optimisation of image acquisition for land cover classification with random forest and MODIS time-series[J]. International Journal of Applied Earth Observations & Geoinformation, 2015, 34: 136-146. DOI:10.1016/j.jag.2014.08.001.
doi: 10.1016/j.jag.2014.08.001
36 Guan H, Li J, Chapman M. Integration of orthoimagery and lidar data for Ooject-based urban thematic mapping using random rorests[J]. International Journal of Remote Sensing, 2013,34(13-14):5166-5186. DOI:10.1080/01431161.2013. 788261.
doi: 10.1080/01431161.2013. 788261
37 Foody G M. Status of land cover classification accuracy assessment[J]. Remote Sensing of Environment, 2002, 80(1): 185-201. DOI:10.1016/S0034-4257(01)00295-4.
doi: 10.1016/S0034-4257(01)00295-4
[1] 姚杰鹏,杨磊库,陈探,宋春桥. 基于Sentinel-1,2和Landsat 8时序影像的鄱阳湖湿地连续变化监测研究[J]. 遥感技术与应用, 2021, 36(4): 760-776.
[2] 梁爽,宫兆宁,赵文吉,关鸿亮,梁亚囡,陆丽,赵雪. 基于多季相Sentinel-2影像的白洋淀湿地信息提取[J]. 遥感技术与应用, 2021, 36(4): 777-790.
[3] 屈炀,袁占良,赵文智,陈学泓,陈家阁. 基于多时序特征和卷积神经网络的农作物分类[J]. 遥感技术与应用, 2021, 36(2): 304-313.
[4] 徐光志,徐涵秋. Sentinel⁃2A MSI 和Landsat 8 OLI两种传感器多光谱信息的交互对比[J]. 遥感技术与应用, 2021, 36(1): 165-175.
[5] 郭利彪,刘桂香,运向军,张勇,孙世贤. 基于数据机理的植被叶面积指数遥感反演研究[J]. 遥感技术与应用, 2020, 35(5): 1047-1056.
[6] 王一帆,徐涵秋. 基于客观阈值与随机森林Gini指标的水体遥感指数对比[J]. 遥感技术与应用, 2020, 35(5): 1089-1098.
[7] 谷祥辉,张英,桑会勇,翟亮,李少军. 基于哨兵2时间序列组合植被指数的作物分类研究[J]. 遥感技术与应用, 2020, 35(3): 702-711.
[8] 董超,赵庚星. 时序数据集构建质量对土地覆盖分类精度的影响研究[J]. 遥感技术与应用, 2020, 35(3): 558-566.
[9] 柴旭荣,李明,周义,王金风,田庆春. 影像的土地覆被快速分类[J]. 遥感技术与应用, 2020, 35(2): 315-325.
[10] 徐玉雯,张浩,陈正超,景海涛. 基于深蓝算法的Sentinel-2数据气溶胶光学厚度反演[J]. 遥感技术与应用, 2020, 35(2): 372-380.
[11] 王一帆,徐涵秋. 利用MODIS EVI时间序列数据分析福建省植被变化(2000~2017年)[J]. 遥感技术与应用, 2020, 35(1): 245-254.
[12] 黄静,王芳,张运. 基于PSI技术的芜湖市地面沉降时空特征分析[J]. 遥感技术与应用, 2019, 34(4): 829-838.
[13] 许文鑫, 周玉科, 梁娟珠. 基于变化点的青藏高原植被时空动态变化研究 [J]. 遥感技术与应用, 2019, 34(3): 667-676.
[14] 吴薇, 张源, 李强子, 黄慧萍. 基于迭代CART算法分层分类的土地覆盖遥感分类[J]. 遥感技术与应用, 2019, 34(1): 68-78.
[15] 汪航,师茁. 基于MODIS时间序列数据的春尺蠖虫害遥感监测方法研究—以新疆巴楚胡杨为例[J]. 遥感技术与应用, 2018, 33(4): 686-695.