Please wait a minute...
img

官方微信

遥感技术与应用  2021, Vol. 36 Issue (2): 400-410    DOI: 10.11873/j.issn.1004-0323.2021.2.0400
农业遥感专栏     
基于Sentinel-2的闪电河流域农作物分类研究
尹燕旻(),贾立()
中国科学院空天信息创新研究院,遥感科学国家重点实验室,北京 100101
Sentinel-2 Study on Crop Mapping of Shandian River Basin based on Images
Yanmin Yin(),Li Jia()
State Key Laboratory of Remote Sensing Science,Institute of Aerospace Information Innovation,Chinese Academy of Sciences,Beijing 100101
 全文: PDF(6402 KB)   HTML
摘要:

以内蒙古闪电河流域为研究区,基于Sentinel2光学遥感影像结合随机森林和支持向量机算法,采用3种方案:基于像元的分类方法、面向对象的分类方法及改进的基于像元分类与面向对象分割相结合的集成方法,对研究区内的农作物进行精细提取。结果表明:①基于随机森林采用基于像元的方法进行分类,所有地类的总体精度为97.8%,Kappa系数为0.974,表明随机森林算法可以有效地进行农作物提取。②改进的基于像元分类与面向对象分割相结合的集成方法分类效果较好,所有地类的总体精度为96.4%,Kappa系数为0.957,该方法充分结合了基于像元和面向对象分类方法的优点,可有效提升闪电河流域的作物分类效果。

关键词: Sentinel-2农作物分类面向对象基于像元随机森林RFSVM    
Abstract:

In this study, Sentinel-2 data combined with Random Forest method (RF) and Support Vector Machine method (SVM) were used to extract crop information in the Shandian River Basin in Inner Mongolia. Three schemes are proposed: pixel-based classification algorithm, object-based classification algorithm and improved integration algorithm based on pixel-based classification and object-based segmentation. Results are as follows: (1) pixel-based classification with RF gets the best extraction accuracy, with an overall accuracy up to 97.8% and Kappa coefficient of 0.974. This result shows that RF has evident advantages in crop extraction. (2) The improved integration algorithm also has good extraction accuracy. The overall accuracy is 96.4%, and kappa coefficient is 0.957. This method fully combines the advantages of pixel-based and object-based classification methods, which effectively improves the crop classification effect in Shandian River region.

Key words: Sentinel-2    Crop extraction    Pixel-based classification    Object-based classification    RF    SVM
收稿日期: 2020-06-22 出版日期: 2021-05-24
ZTFLH:  TP79  
基金资助: 中国科学院战略性先导科技专项(XDA19030203)
通讯作者: 贾立     E-mail: yinym@radi.ac.cn;jiali@radi.ac.cn
作者简介: 尹燕旻(1994-),女,山东泰安人,硕士研究生,主要从事农作物分类研究。E?mail:yinym@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  
尹燕旻
贾立

引用本文:

尹燕旻,贾立. 基于Sentinel-2的闪电河流域农作物分类研究[J]. 遥感技术与应用, 2021, 36(2): 400-410.

Yanmin Yin,Li Jia. Sentinel-2 Study on Crop Mapping of Shandian River Basin based on Images. Remote Sensing Technology and Application, 2021, 36(2): 400-410.

链接本文:

http://www.rsta.ac.cn/CN/10.11873/j.issn.1004-0323.2021.2.0400        http://www.rsta.ac.cn/CN/Y2021/V36/I2/400

  图1闪电河流域实验区区位图
获取日期获取卫星波段数影像质量作物生长期
2018-08-01S2A13无云

胡萝卜、土豆:块茎膨大期

莜麦:抽穗期

2018-09-05S2B13无云

胡萝卜、土豆:成熟期

莜麦:收获期

表1  闪电河流域研究区Sentinel2遥感影像信息
方案尺度玉米土豆胡萝卜莜麦林地草地建设用地荒地
方案一像元9 67234 75219 50111 70922 6498 71610 5603 679
252641 0185774325155841 531285
方案二50100423235175152321400107
10046193111866319015349
表2  闪电河研究区农作物分类样本统计表
图 2  闪电河流域作物分类实验方案流程图
方案子实验详细信息
1a基于像元的随机森林分类
b基于像元SVM分类
2c25、50、100分割尺度下基于对象的随机森林分类
d25、50、100分割尺度下基于对象的SVM分类
3e依据方案1中优选像元分类结果与方案2中不同尺度分割结果,进行对象内聚合
表3  闪电河研究区作物分类实验方案信息
特征变量简称说明
光谱特征BB2、B3、B4、B5、B6、B7、B8、B8a、B11、B12
指数特征NDVI(B8a-B4)/(B8a+B4)
NDWI(B3-B8a)/( B3+B8a)
GLCM-A角二矩阵
GLCM-cor对比度
纹理特征GLCM-con相关性
GLCM-E
GLCM-dis非相似性
GLCM-H同质度
表4  作物分类像元特征集描述
统计特征简称像素特征类别说明
均值Mean光谱特征求一个对象内所有像素的像素特征的均值
植被指数
水体指数
纹理特征
标准差Std光谱特征求一个对象内所有像元的像元特征的标准差
植被指数
水体指数
纹理特征
贡献率Ratio光谱特征求同一期影像内,一个对象的某一光谱特征平均值与所有光谱特征平均值总和的比率
纹理特征求同一期影像内,一个对象的某一纹理特征平均值与所有纹理特征平均值总和的比率
表5  作物分类对象特征集描述
图3  特征重要性分布
重要性得分0.00~0.020.02~0.030.03~0.050.05~0.070.07~0.09
特征权重12468
表6  对象分割输入特征权重对应表
图4  不同方案精度统计图(横坐标A、B、C、D、E、F、G分别表示玉米、土豆、胡萝卜、莜麦、林地、草地、建设用地、荒地)
图5  闪电河流域研究区内不同分类方案局部地区作物分布图(2018年)
图6  闪电河流域研究区内不同分类方案作物分布图(2018年)
方案一方案二方案三
类别子实验a子实验b子实验c子实验d子实验e
PA/%UA/%PA/%UA/%PA/%UA/%PA/%UA/%PA/%UA/%
玉米92.2899.8379.5796.2083.9395.9266.0788.1083.93100
土豆99.1197.1796.9492.1196.4593.1489.3476.5297.9793.24
胡萝卜99.2298.6594.1194.7895.9095.1274.5981.2596.7298.33
莜麦99.5899.5391.2694.4697.9896.0466.6784.6210098.02
树林99.6795.5797.5786.0293.1093.1086.2174.2698.8586.87
草地87.0898.0054.5776.3388.3389.0866.6770.1884.1796.19
居民地99.1899.0397.3997.3499.0298.0510097.7610099.03
裸土98.4399.6096.9997.6391.0710094.6410098.21100
总精度%97.8291.5595.1184.7496.35
Kappa系数0.9740.8970.9410.8150.957
表7  闪电河流域研究区内作物分类精度统计
1 Chen Zhongxin, Ren Jjianqiang, Tang Huajun,et al. Progress and Perspectives on Agricultural Remote Sensing Research and Applications in China[J]. Journal of Remote Sensing, 2016,20(5): 748–767.
1 陈仲新,任建强,唐华俊,等.农业遥感研究应用进展与展望[J].遥感学报,2016,20(5):748-767.
2 Zhao Chunjiang.Advances of Research and Application in Remote Sensing for Agriculture[J]. Transactions of the Chinese Society for Agricultural Machinery,2014,45(12):277-293.
2 赵春江.农业遥感研究与应用进展[J].农业机械学报, 2014,45(12):277-293.
3 Wu Bingfang, Zhang Miao, Zeng Hongwei,et al. Agricultural Monitoring and Early Warning in the Era of Big Data[J]. Journal of Remote Sensing, 2016,20(5):1027-1037.
3 吴炳方,张淼,曾红伟,等. 大数据时代的农情监测与预警[J]. 遥感学报,2016,20(5):1027-1037.
4 Meng Jihua,Wu Bingfang,Du Xin,et al. A Review and Outlook of Applying Remote Sensing to Precision Agriculture[J] .Remote Sensing for Land and Resources,2011,23(3):1-7.
4 蒙继华,吴炳方,杜鑫,等. 遥感在精准农业中的应用进展及展望[J].国土资源遥感,2011,23 (3): 1-7.
5 Patil P, Murali K G, Gumma, et al. A Review of the Available Land Cover and Cropland Maps for South Asia[J]. Agriculture, 2018,8(7):111. doi:10.3390/agriculture 8070111.
doi: 10.3390/agriculture 8070111
6 Song Qian,Zhou Qingbo,Wu Wenbin,et al.Recent Progresses in Research of Integrating Multi-source Remote Sensing Data for Crop Mapping[J].Scientia Agricultura Sinica,2015,48(6):1122-1135.
6 宋茜,周清波,吴文斌,等. 农作物遥感识别中的多源数据融合研究进展[J].中国农业科学,2015,48(6): 1122-1135.
7 Liu Zhe,Liu Diyou,Zhu Dehai,et al. Review on Crop Type Fine Identification and Automatic Mapping Using Remote Sensing [J].Transactions of the Chinese Society for Agricultural Machinery, 2018,49(12):1-12.
7 刘哲,刘帝佑,朱德海,等.作物遥感精细识别与自动制图研究进展与展望[J]. 农业机械学报,2018,49(12):1-12.
8 Liu Bingjing,Yang Yanzhao,Li Yi. Quantitative Analysis of Land Use Structure Characteristics over the Farming——Pastoral Zone in the West Liaohe River basin,Northern China[J]. Journal of Arid Land Resources and Environment, 2018,32(6):64-71.
8 刘冰晶, 杨艳昭, 李依. 北方农牧交错带土地利用结构特征定量研究——以西辽河流域为例[J]. 干旱区资源与环境, 2018, 32(6):64-71.
9 Han Tao,Pan Jianjun,Zhang Peiyu,et al.Study on Differences between Sentinel-2A and Landsat 8 Images in Rape identification[J]. Remote Sensing Technology and Application , 2018,33(5):890-899.
9 韩涛,潘剑君,张培育,等. Sentinel-2 与Landsat 8影像在油菜识别中的差异性研究[J].遥感技术与应用,2018,33(5):890-899.
10 Chen Mingye,Chen Lei,Zhou Xun.A Remote Sensing Study of Spatio-temporal Changes of Ecological Environment of Shandian River[J] .Remote Sensing for Land and Resources,2017,29(4):166 -172.
10 陈明叶,陈磊, 周询. 闪电河上游生态环境时空变化遥感研究[J].国土资源遥感,2017,29(4):166-172.
11 Zhao Y, Potgieter A B, Zhang Miao, et al. Predicting Wheat Yield at the Field Scale by Combining High-resolution Sentinel-2 Satellite Imagery and Crop Modelling[J].Remote Sensing, 2020,12(6):1024. doi:10.3390/rs12061024.
doi: 10.3390/rs12061024
12 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,2012,120:25-36.doi: 10.1016/j.rse.2011.11.026.
doi: 10.1016/j.rse.2011.11.026
13 Li Qiliang,Fan Jinlong,Xu Qi,Liu Shaojie,et al. Ground Sample Survey based on GPS Photos Processing System[J]. Journal of Geomatics, 2019(3):113-116.李启亮, 范锦龙, 许淇,等. 基于GPS照片数据处理系统的地面样方调查[J]. 测绘地理信息, 2019(3):113-116.
14 Zhang X, Sun Y L, Shang K, et al. Crop Classification based on Feature Band Set Construction and Object-Oriented Approach Using Hyperspectral Images[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing,2016,9(9):4117-4128. doi:10.1109/JSTARS.2016. 2577339.
doi: 10.1109/JSTARS.2016. 2577339
15 Zhou Zhuang,Li Shengyang,Zhang Kang,et al. Crop Mapping Using Remotely Sensed Spectral and Context Features based on CNN[J].Remote Sensing Technology and Application,2019,34(4):694-703.
15 周壮,李盛阳,张康,等.基于CNN和农作物光谱纹理特征进行作物分布制图[J].遥感技术与应用,2019,34(4):694-703.
16 Song Q, Zhou Q B, Wu W B, et al. Mapping Regional Cropping Patterns by using GF-1 WFV Sensor Data[J]. Journal of Integrative Agriculture,2017,16(2):337-347.doi:10.1016/S2095-3119(16)61392-8.
doi: 10.1016/S2095-3119(16)61392-8
17 Zhang Lei, Gong Zhaoning, Wang Qiwei, et al. Wetland Mapping of Yellow River Delta Wetlands based on Multi-feature Optimization of Sentinel-2 Images[J]. Journal of Remote Sensing,2019,23(2):313-326.
17 张磊,宫兆宁,王启为,等.Sentinel-2影像多特征优选的黄河三角洲湿地信息提取[J].遥感学报,2019,23(2):313-326.
18 Du S H,Guo Z,Wang W Y,et al. A Comparative Study of the Segmentation of Weighted Aggregation and Multiresolution Segmentation[J]. Giscience & Remote sensing,2016,53(5):651-670.
19 Shi Peirong,Chen Yongfu,Liu Hua, et al.Parameters of Multi-segmentation based on Segmentation Evalution Function[J]. Remote Sensing Technology and Application, 2018,33(4):628-637.
19 施佩荣,陈永富,刘华,等,基于分割评价函数的多尺度分割参数的选择[J].遥感技术与应用,2018,33(4):628-637.
20 Ma L,Li M C,Ma X X,et al. A Review of Supervised Object-based Land-cover Image Classification[J]. Elsevier Bes-loten Vennootschap,2017,130.doi:10.1016/j.isprsjprs.2017.06.001.
doi: 10.1016/j.isprsjprs.2017.06.001
21 Breiman L.Random Forests[J]. Machine Learning, 2001, 45(1):5-32.doi:10.1023/A:1010933404324.
doi: 10.1023/A:1010933404324
22 Belgiu M, Drăguţ L. Random Forest in Remote Sensing: A Review of Applications and Future Directions[J]. ISPRS Journal of Photogrammetry and Remote Sensing,2016,114:24-31.doi:10.1016/j.isprsjprs.2016.01.011.
doi: 10.1016/j.isprsjprs.2016.01.011
23 Mountrakis G, Im J, Ogole C. Support Vector Machines in Remote Sensing: A Review[J].ISPRS Journal of Photogrammetry and Remote Sensing,2010,66(3):247-259. doi:10.1016/j.isprsjprs.2010.11.001.
doi: 10.1016/j.isprsjprs.2010.11.001
24 Xiong J, Thenkabail P S, Tilton J C,et al. Nominal 30 m Cropland Extent Map of Continental Africa by Integrating Pixel-Based and Object-based Algorithms Using Sentinel-2 and Landsat 8 Data on Google Earth Engine[J]. Remote Sensing,2017,9(10):1065.doi: 10.3390/rs9101065.
doi: 10.3390/rs9101065
25 Li Xiang,Liu Kai,Zhu Yuanhui,et al.Study on Mangrove Species Classification based on ZY-3 Image[J].Remote Sensing Technologyand Application,2018,33(2):360-369.
25 李想,刘凯,朱远辉,等.基于资源三号影像的红树林物种分类研究[J].遥感技术与应用,2018,33(2):360-369.
26 Carolin S, Anne-Laure B, Achim Z, et al. Bias in Random Forest Variable Importance Measures: Illustrations, Sources and a Solution[J]. BMC Bioinformatics,2007,8:25. doi:10.1186/1471-2105-8-25.
doi: 10.1186/1471-2105-8-25
[1] 胡乃勋,陈涛,甄娜,牛瑞卿. 基于卷积神经网络的面向对象露天采场提取[J]. 遥感技术与应用, 2021, 36(2): 265-274.
[2] 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.
[3] 陈彦四,黄春林,侯金亮,韩伟孝,冯娅娅,李翔华,王静. 基于多时相Sentinel-2影像的黑河中游玉米种植面积提取研究[J]. 遥感技术与应用, 2021, 36(2): 324-331.
[4] 杨福芹,冯海宽,李振海,潘洁晨,谢瑞. 苹果叶片氮含量高光谱反演方法对比[J]. 遥感技术与应用, 2021, 36(2): 353-361.
[5] 王晨丞,王永前,王利花. 基于SAR纹理信息的农作物识别研究——以农安县为例[J]. 遥感技术与应用, 2021, 36(2): 372-380.
[6] 王崇阳,田昕. 基于GF⁃1 PMS数据的森林覆盖变化检测[J]. 遥感技术与应用, 2021, 36(1): 208-216.
[7] 付甜梦,张丽,陈博伟,闫敏. 基于GEE平台的海岛地表覆盖提取及变化监测—以苏拉威西岛为例[J]. 遥感技术与应用, 2021, 36(1): 55-64.
[8] 马腾耀,肖鹏峰,张学良,马威,郭金金. 基于特征优选的GF-3全极化数据积雪识别[J]. 遥感技术与应用, 2020, 35(6): 1292-1302.
[9] 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.
[10] 郑琪,邸苏闯,潘兴瑶,刘洪禄,朱永华,张岑,周星. 基于Rapid Eye数据的北京生态涵养区土地利用分类及变化研究[J]. 遥感技术与应用, 2020, 35(5): 1118-1126.
[11] 牟昱璇,邬明权,牛铮,黄文江,杨尽. 南方地区复杂条件下的耕地面积遥感提取方法[J]. 遥感技术与应用, 2020, 35(5): 1127-1135.
[12] 王明,刘正佳,陈元琰. 基于Sentinel-2波段/产品的图像云检测效果对比研究[J]. 遥感技术与应用, 2020, 35(5): 1167-1177.
[13] 李宏达,高小红,汤敏. 基于CNN的不同空间分辨率影像土地覆被分类研究[J]. 遥感技术与应用, 2020, 35(4): 749-758.
[14] 杨瑞,祁元,苏阳. 深度学习U-Net方法及其在高分辨卫星影像分类中的应用[J]. 遥感技术与应用, 2020, 35(4): 767-774.
[15] 连喜红,祁元,王宏伟,张金龙,杨瑞. 基于面向对象的青海湖环湖区居民地信息自动化提取[J]. 遥感技术与应用, 2020, 35(4): 775-785.