Please wait a minute...
img

官方微信

遥感技术与应用  2020, Vol. 35 Issue (3): 702-711    DOI: 10.11873/j.issn.1004-0323.2020.3.0702
遥感应用     
基于哨兵2时间序列组合植被指数的作物分类研究
谷祥辉1(),张英2,桑会勇2,翟亮2(),李少军3
1.山东科技大学 测绘科学与工程学院,山东 青岛 266590
2.中国测绘科学研究院, 北京 100830
3.新疆维吾尔自治区测绘科学研究院, 新疆 乌鲁木齐 830002
Research on Crop Classification Method based on Sentinel-2 Time Series Combined Vegetation Index
Xianghui Gu1(),Ying Zhang2,Huiyong Sang2,Liang Zhai2(),Shaojun Li3
1.College of Geomatics, Shandong University of Science and Technology, Qingdao 266590, China
2.Chinese Academy of Surveying and Mapping, Beijing 100830, China
3.Xinjiang Academy of Surveying and Mapping, Xinjiang Uygur Autonomous Region Urumqi, Urumqi 830002,China
 全文: PDF(5926 KB)   HTML
摘要:

时间序列是一种常用的物候研究方法。为充分利用哨兵2数据在红边波段的丰富信息,本文利用多种植被指数组合成时间序列进行作物分类。将NDVI、EVI、红边NDVI三种植被指数进行组合,构建时序植被指数图像,然后使用支持向量机、随机森林、CART决策树和最大似然4种不同的算法对四种作物、三种林草、裸露地表、水体进行分类。原始分类结果中,总体精度最高的随机森林为87.92%,最低的最大似然为80.07%,在分类细节上,随机森林和支持向量机的边界最清晰,4种分类结果中,农作物的分类精度均高于其他地类,仅次于水体的精度,误差主要来自三种林草的混分,表明时间序列组合植被指数用于农作物分类是可行的。

关键词: 植被指数时间序列遥感农作物分类哨兵2    
Abstract:

Time series is a widely used phenological research method. A new time series vegetation indices which takes full advantage of the red edge information of Sentinel 2 data were used for crop classification to improve the classification accuracy. The NDVI, EVI, and red edge NDVI were combined to construct a time series vegetation index image. Then, four different algorithms (support vector machine, random forest, CART decision tree and maximum likelihood) were used to classify four crops, three forest grasses, bare land, and water bodies. Among the original classification results, the random forest with the highest overall accuracy is 87.92%, and the maximum likelihood with the lowest overall accuracy is 80.07%. In the classification details, the boundaries of random forest and support vector machine are the clearest. Among the four classification results, the classification accuracy of crops is higher than other land types, just smaller than water body. The error mainly comes from the mixture of three forests. It indicates that the time series combined vegetation index is feasible and accurate for crop classification.

Key words: Vegetation index    Time series    Remote sensing    Crop classification    Sentinel 2
收稿日期: 2019-07-01 出版日期: 2020-07-10
ZTFLH:  TP79  
基金资助: 中国测绘科学研究院基本科研业务费项目(7771728);自然资源部地球观测与时空信息科学重点实验室开放基金(AR191902)
通讯作者: 翟亮     E-mail: 1090441654@qq.com;zhailiang@casm.ac.cn
作者简介: 谷祥辉(1993-),男,山东枣庄人,硕士研究生,主要从事地理国情遥感监测、定量遥感等方面的研究。E?mail:1090441654@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  
谷祥辉
张英
桑会勇
翟亮
李少军

引用本文:

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.

Xianghui Gu,Ying Zhang,Huiyong Sang,Liang Zhai,Shaojun Li. Research on Crop Classification Method based on Sentinel-2 Time Series Combined Vegetation Index. Remote Sensing Technology and Application, 2020, 35(3): 702-711.

链接本文:

http://www.rsta.ac.cn/CN/10.11873/j.issn.1004-0323.2020.3.0702        http://www.rsta.ac.cn/CN/Y2020/V35/I3/702

图 1  研究区概况
波段S2AS2B空间分辨率/m备注
中心波长/nm波段宽度/nm中心波长/nm波段宽度/nm
2492.498492.19810
3559.84555946
4664.638664.939
8832.8145832.9133
5704.119703.82020窄波段 红边
6740.518739.118
7782.828779.728
8a864.73386432
111 613.71431 610.414120
122 202.42422 185.7238
1442.727442.24560
9945.126943.227
101 373.5751 376.976
表 1  哨兵2数据空间与光谱分辨率
获取日期儒略日*农作物所处发育期
冬小麦夏玉米春玉米大豆花生
2017/2/2758 d返青
2017/3/968 d起身
2017/3/2988 d起身
2017/4/18108 d拔节播种
2017/5/18138 d开花出苗
2017/5/28148 d乳熟
2017/6/7158 d成熟抽雄
2017/6/27178 d收割播种吐丝播种播种
2017/7/7188 d出苗出苗出苗
2017/7/12193 d拔节乳熟开花
2017/8/6218 d抽雄花芽下针
2017/8/16228 d吐丝成熟结荚结荚
2017/9/20263 d乳熟收割成熟收割
2017/10/20293 d播种收割收割
2017/10/30303 d出苗
表 2  主要农作物发育期
指数计算方法波幅宽度要求
NDVIρNIR-ρR)/(ρNIR+ρR)
EVI2.5*(ρNIR-ρR)/(ρNIR+6ρR-7.5ρB+1)
NDVI705ρ750-ρ705)/(ρ750+ρ705)
表3  3种植被指数及公式
图2  林草类时序曲线
图3  农作物时序曲线
图4  样本区位置
类别指标分类方法
SVMRFCARTMLC
冬小麦玉米大豆制图精度89.6889.5484.7379.84
用户精度90.2489.8889.8992.84
冬小麦花生制图精度88.3988.2179.3083.00
用户精度95.6093.1986.7493.70
春玉米制图精度93.1291.4889.2991.06
用户精度82.8887.6179.5667.65
夏玉米制图精度87.7987.4487.0479.04
用户精度91.1792.0487.0395.26
林地制图精度85.0377.7076.0877.70
用户精度67.6167.4962.1643.91
天然草地制图精度88.8687.7183.8576.21
用户精度91.2493.7292.1594.18
牧草制图精度82.3481.7672.7686.26
用户精度78.0270.7161.6440.54
水体制图精度98.2699.2199.3090.64
用户精度99.9399.9997.58100.00
裸露地表制图精度79.6385.5479.9975.51
用户精度87.3881.9078.1584.56
总体精度87.90%87.92%84.25%80.07%
Kappa系数0.857 90.858 10.815 40.769 3
表 4  原始分类结果分类精度
图5  分类细节
指标分类方法
SVMRFCARTMLC
总体精度88.37%88.52%85.89%81.03%
Kappa系数0.863 30.865 00.834 40.780 1
表 5  消除孤岛后的精度
图6  原始分类结果
1 Wang Xue, Li Xiubin, Xin Liangjie, et al. Driving Factors on Winter Wheat Abandonment by Farmers in North China Plain and Its Policy Implication[J]. Transactions of the Chinese Society of Agricultural Engineering, 2018, 34(9):248-257.
1 王学,李秀彬,辛良杰,等.华北平原农户主动退耕冬小麦的影响因素模型分析及政策启示[J].农业工程学报,2018,34(9):248-257.
2 Li Shengfa, Li Xiubin. Economic Characteristics and the Mechanism of Farmland Marginalization in Mountainous Areas of China[J]. Acta Geographica Sinica,2018, 73(5):803-817.
2 李升发,李秀彬.中国山区耕地利用边际化表现及其机理[J].地理学报,2018,73(5):803-817.
3 Li Shengfa, Li Xiubin, Xin Liangjie, et al. Extent and Distribution of Cropland Abandonment in Chinese Mountainous Areas[J]. Resources Science,2017,39(10):1801-1811.
3 李升发,李秀彬,辛良杰,等.中国山区耕地撂荒程度及空间分布——基于全国山区抽样调查结果[J].资源科学,2017,39(10):1801-1811.
4 Zhang Xiwang, Liu Jianfeng, Qin Fen, et al. A Review of Remote Sensing Application in Crop Type Discrimination[J]. Chinese Agricultural Science Bulletin,2014,30(33):278-285.
4 张喜旺,刘剑锋,秦奋,等.作物类型遥感识别研究进展[J].中国农学通报, 2014,30(33):278-285.
5 Su Tengfei, Liu Quanming, Su Xiuchuan. Research on the Remote Sensing Classification of Crops based on Multiple Vegetation Index Time Series and Machine Learning[J]. Jiangsu Agricultural Science, 2017, 45(16):219-224.
5 苏腾飞,刘全明,苏秀川.基于多种植被指数时间序列与机器学习的作物遥感分类研究[J].江苏农业科学,2017,45(16): 219-224.
6 Liu Jianfeng, Zhang Xiwang. Research on Remote Sensing Identification Method of Maize based on Seasonal Rhythms and Spectral Characteristics[J].Remote Sensing Technology and Application, 2016,31(6): 1131-1139.
6 刘剑锋,张喜旺.基于光谱和时相特征的夏玉米遥感识别[J].遥感技术与应用,2016,31(6): 1131-1139.
7 Ping Yuepeng, Zang Shuying. Crop Identification based on MODIS NDVI Time-series Data and Phenological Characteristics[J]. Journal of Natural Resources, 2016,31(3):503-513.
7 平跃鹏,臧淑英.基于MODIS时间序列及物候特征的农作物分类[J].自然资源学报,2016,31(3):503-513.
8 Liu Yizhu, Wu Wenbin, Li Zhaoliang, et al. Extracting Irrigated Cropland Spatial Distribution in China based on time-series NDVI[J]. Transactions of the Chinese Society of Agricultural Engineering,2017,33(22):276-284.
8 刘逸竹,吴文斌,李召良,等.基于时间序列NDVI的灌溉耕地空间分布提取[J].农业工程学报,2017,33(22):276-284.
9 Hu Wen, Liu Huanjun, Zhang Xinle, et al. Extraction of Cultivated Land Scope based on MODIS Vegetation Index Time-series Data—A Case Study of Heilongjiang Province[J]. Southern Journal of Agriculture,2015,46(8):1533-1537.
9 胡文,刘焕军,张新乐,等.基于MODIS植被指数时间序列的耕地范围提取—以黑龙江省为例[J]. 南方农业学报,2015,46(8):1533-1537.
10 Guo Yushan, Liu Qingsheng, Liu Gaohuan,et al. Extraction of Main Crops in Yellow River Delta based on MODIS NDVI Time Series [J]. Journal of Natural Resources,2017,32 (10):1808-1818.
10 郭昱杉,刘庆生,刘高焕,等.基于MODIS时序NDVI主要农作物种植信息提取研究[J].自然资源学报, 2017, 32(10): 1808-1818.
11 Yang Xiaoxue, Zhang Xiangping, Jiang Dong. Extraction of Multi-Crop Planting Areas from MODIS Data[J].Resource Science,2004,26(6):17-22.
11 杨小唤,张香平,江东.基于MODIS时序NDVI特征值提取多作物播种面积的方法[J]. 资源科学,2004,26(6):17-22.
12 Yu Lifeng, Tuya Wulan, Deji Wuyun,et al. Study on Extraction of Arable Land Area based on Texture Features and MODIS-NDVI Time Series[J]. Chinese Journal of Agricultural Resources and Regional Planning, 2018,39(11):169-177.
12 于利峰,乌兰吐雅,乌云德吉,等.基于纹理特征与MODIS_NDVI时间序列的耕地面积提取研究[J].中国农业资源与区划,2018,39(11):169-177.
13 Liu Xinjie, Wei Yunxia, Jiao Quanjun,et al. Growth Monitoring and Yield Prediction of Winter Wheat based on Time-series Quantitative Remote Sensing Data[J]. Remote Sensing Technology and Application, 2019, 34(4): 756-765.
13 刘新杰,魏云霞,焦全军,等. 基于时序定量遥感的冬小麦长势监测与估产研究[J]. 遥感技术与应用, 2019, 34(4): 756-765.
14 Pan Yaozhong, Li Le, Zhang Jinshui,et al. 2011. Crop Area Estimation based on MODIS-EVI Time Series According to Distinct Characteristics of Key Phenology Phases: A Case Study of Winter Wheat Area Estimation in Small-scale Area[J]. Journal of Remote Sensing, 2011,15(3): 578-594.
14 潘耀忠,李乐,张锦水,等.基于典型物候特征的MODIS-EVI时间序列数据农作物种植面积提取方法——小区域冬小麦实验研究[J].遥感学报,2011,15(3):578-594.
15 Wang Xue, Li Xiubin, Tan Minghong, et al. Remote Sensing Monitoring of Changes in Winter Wheat Area in North China Plain from 2001 to 2011[J]. Transactions of the Chinese Society of Agricultural Engineering,2015,31(8):190-199.
15 王学,李秀彬,谈明洪,等.华北平原2001~2011年冬小麦播种面积变化遥感监测[J].农业工程学报,2015,31(8):190-199.
16 Gao Yingbo, Liu Qinhuo, Li Jing, et al. A New Methodology for Extracting Multiple Cropping Land based on Distinguishing Characteristic Phases of Time Series Vegetation Index [J]. Remote Sensing Technology and Applications,2015,30(3):431-438.
16 高应波,柳钦火,李静,等.基于时序植被指数特征时相识别的多熟制耕地提取新方法[J].遥感技术与应用,2015,30(3):431-438.
17 Yu Fengming, Zhuo Yi, Bao Yuhai. BP Neural Network Used in Classification of Land Cover[J]. Science of Surveying and Mapping,2008,33(Sup.3):203-204.
17 于凤鸣,卓义,包玉海.基于BP神经网络和MODIS-EVI时间序列的土地覆被分类[J].测绘科学,2008,33(增刊3):203-204.
18 Zhang Rongqun, Wang Sheng'an, Gao Wanlin, et al. Remote-sensing Classification Method of County-level Agricultural Crops Using Time-series NDVI[J]. Transactions of the Chinese Society for Agricultural Machinery,2015,46(Sup.1):246-252.
18 张荣群,王盛安,高万林,等.基于时序植被指数的县域作物遥感分类方法研究[J].农业机械学报,2015,46(增刊1):246-252.
19 Shi Feifei, Gao Xiaohong, Yang Lingyu, et al. Research on Typical Crop Classification based on HJ-1A Hyperspectral Data in The Huangshui River Basin[J]. Remote Sensing Technology and Applications,2017,32(2):206-217.
19 史飞飞,高小红,杨灵玉,等.基于HJ_1A高光谱遥感数据的湟水流域典型农作物分类研究[J].遥感技术与应用,2017,32(2):206-217.
20 Li Xiaohui, Wang Hong, Li Xiaobing, et al. Study on Crops Remote Sensing Classification based on Multi-temporal Landsat 8 OLI Image[J]. Remote Sensing Technology and Applications,2019,34(2):389-397.
20 李晓慧,王宏,李晓兵,等.基于多时相Landsat 8 OLI影像的农作物遥感分类研究[J].遥感技术与应用,2019,34(2):389-397.
21 Hao Pengyu. Crop Classification Using Time Series Remote Sensing Data[D]. Beijing: University of Chinese Academy of Sciences (Institute of Remote Sensing and Digital Earth, Chinese Academy of Sciences),2017.郝鹏宇.基于多时相遥感数据的农作物分类研究[D].北京:中国科学院大学(中国科学院遥感与数字地球研究所),2017.
22 Xiao Guofeng, Zhu Xiufang, Hou Chenyao, et al. Extraction and Analysis of Abandoned Farmland: A Case Study of Qingyun and Wudi counties in Shandong Province[J]. Acta Geographica Sinica,2018,73(9):1658-1673.
22 肖国峰,朱秀芳,侯陈瑶,等.撂荒耕地的提取与分析—以山东省庆云县和无棣县为例[J].地理学报,2018,73(9):1658-1673.
23 Drusch M, Del Bello U, Carlier S, et al. Sentinel-2: ESA’s Optical High-Resolution Mission for GMES Operational Services[J]. Remote Sensing of Environment,2012,120(2018):25-36..
24 Lambert Marie-Julie, Pierre C. Sibiry Traoré,et al. Estimating Smallholder Crops Production at Village Level from Sentinel-2time Series in Mali's Cotton Belt[J]. Remote Sensing of Environment, 2018, 216(2018): 647-657. .
25 Wang Lijun, Guo Yan, He Jia, et al. Classification Method by Fusion of Decision Tree and SVM based on Sentinel-2A Image[J]. Transactions of the Chinese Society for Agricultural Machinery,2018,49(9):146-153.
25 王利军,郭燕,贺佳,等.基于决策树和SVM的Sentinel-2A影像作物提取方法[J],农业机械学报,2018,49(9):146-153.
26 Vrieling A, Meroni M, Darvishzadeh R, et al. Vegetation Phenology from Sentinel-2 and Field Cameras for a Dutch Barrier Island[J]. Remote Sensing of Environment, 2018,215: 517–529. .
27 Anatoly G, Mark N. Spectral Reflectance Changes Associated with Autumn Senescence of Aesculus Hippocastanum L. and Acer Platanoides L. Leaves. Spectral Features and Relation to Chlorophyll Estimation[J]. Journal of Plant Physiology,1994, 143(3):286-292. doi:10.1016/S0176-1617(11)81633-0.
doi: 10.1016/S0176-1617(11)81633-0
28 Daniel A, John A. Relationships between Leaf Pigment Content and Spectral Reflectance Across a Wide Range of Species, Leaf Structures and Developmental Stages[J]. Remote Sensing of Environment,2002, 81(2): 337-354. doi:10.1016/ S0034-4257(02)00010-X.
doi: 10.1016/ S0034-4257(02)00010-X
29 Wang Limin, Liu Jia, Yang Lingbo, et al. Impact of Short Infrared Wave band on Identification Accuracy of Corn and Soybean Area[J]. Transactions of the Chinese Society of Agricultural Engineering,2016,32(19):169-178.
29 王利民,刘佳,杨玲波.短波红外波段对玉米大豆种植面积识别精度的影响[J].农业工程学报,2016,32(19):169-178.
[1] 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.
[2] 唐希颖,崔耀平,李楠,付一鸣,刘小燕,闰亚迪. 2000~2015年北京市土地利用强度及其辐射反馈评估[J]. 遥感技术与应用, 2020, 35(3): 587-595.
[3] 史姝姝,匡文慧,董斯齐. 21世纪以来西安城乡梯度土地覆盖变化及对城市热岛影响时空特征[J]. 遥感技术与应用, 2020, 35(3): 537-547.
[4] 李净,温松楠. 基于3种机器学习法的太阳辐射模拟研究[J]. 遥感技术与应用, 2020, 35(3): 615-622.
[5] 王红岩,汪晓帆,高亮,李强子,赵龙才,杜鑫,张源. 基于季相变化特征的撂荒地遥感提取方法研究[J]. 遥感技术与应用, 2020, 35(3): 596-605.
[6] 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.
[7] 虞瑶,苏红军,姚文静. 基于Boosting的高光谱遥感切空间协同表示集成学习方法[J]. 遥感技术与应用, 2020, 35(3): 634-644.
[8] 马彩虹,关琳琳,陈甫,王大成,刘建波. 基于内容的遥感图像变化信息检索概念模型设计[J]. 遥感技术与应用, 2020, 35(3): 685-693.
[9] 岳昂,曾庆伟,王怀警. 于桥水库蓝藻水华遥感长时序监测研究[J]. 遥感技术与应用, 2020, 35(3): 694-701.
[10] 寇勇,王宁练,陈安安,刘凯. 1993~2016年喜马拉雅山西段杰纳布流域冰川变化遥感监测[J]. 遥感技术与应用, 2020, 35(3): 712-722.
[11] 张平,孙强强,张亚萍,孙丹峰,刘顺喜. 基于宽波段遥感光谱混合分解的干旱区土地退化监测评价理论和技术方法探讨[J]. 遥感技术与应用, 2020, 35(2): 497-508.
[12] 程娟,肖青,闻建光,唐勇,游冬琴,卞尊健,吴胜标,郝大磊,钟守熠. 地物波谱数据库应用方法及遥感应用现状[J]. 遥感技术与应用, 2020, 35(2): 267-286.
[13] 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.
[14] 谭美宝,冉有华,苏阳,李新,杜得彦,廉耀康. 黑河流域2001~2017年植被变化特征及其可延续性评价[J]. 遥感技术与应用, 2020, 35(2): 335-344.
[15] 纪甫江,蒙继华,方慧婷. 基于CASA-WOFOST耦合模型的大豆单产遥感估算研究[J]. 遥感技术与应用, 2020, 35(2): 406-415.