Please wait a minute...


遥感技术与应用  2020, Vol. 35 Issue (3): 558-566    DOI: 10.11873/j.issn.1004-0323.2020.3.0558
1.山东农业大学信息科学与技术学院,山东 泰安 271018
2.山东农业大学资源与环境学院,土肥资源高效利用国家工程实验室,山东 泰安 2710181
Influence of Time Series Data Quality on Land Cover Classification Accuracy
Chao Dong1,2(),Gengxing Zhao2()
1.College of Information Science And Engineering, Shandong Agricultural University, Tai'an 271018, China
2.College of Resources and Environment, National Engineering Laboratory for Efficient Utilization of Soil and Fertilizer Resources, Shandong Agricultural University, Tai'an 271018, China
 全文: PDF(5102 KB)   HTML

研究通过对MODIS双星数据组合、线性插值和HANTS平滑方法来提升时序数据集质量,采用随机森林的方法分类,对分类结果精度评定以分析时序数据集构建质量对分类精度的影响。结果表明:双星数据有利于提高时序数据集的时间分辨率,精确刻划覆盖变化,为后续处理提供基础;线性插值可改善像元点的质量,降低云、雨因素影响;HANTS平滑能移除异常值,平滑数据,突出曲线特征,降低分类复杂度。改进质量后的时序数据集,分类总体精度从84.32%提高至90.75%,Kappa系数从0.798 6提高至0.881 6。总之,使用时序数据进行土地覆盖分类时,应以消除异常值,真实反映地表覆盖物候特征为目的提高时序数据集的质量,从而提高分类精度。

关键词: MODIS时间序列HANTS精度评价随机森林    

This paper improved the quality of time series data sets through three


double star data combination of MODIS, linear interpolation and HANTS smoothing. In this study, we used random forest classification and analyzed the impact of the quality of time series dataset construction on classification accuracy though evaluating the accuracy of classification results. Results showed that the double-star data was beneficial to improve the temporal resolution of time series dataset, accurately depict the coverage change, and provide the basis for subsequent processing; linear interpolation could improve the quality of pixel points and reduce the influence of cloud and rain factors; HANTS smoothing could remove outliers, smooth data, highlight curve features, and reduce classification complexity. After improving the quality of the time series data set, the overall classification accuracy increased from 84.32% to 90.75%, and the Kappa coefficient increased from 79.86% to 88.16%. In a word, when using time series data for land cover classification, the quality of the time series data set should be improved to eliminate the outliers and truly reflect the surface covering phenological features, and the classification accuracy of the results should be improved.

Key words: MODIS    Time series    HANTS    Accuracy assessment    Random forest
收稿日期: 2019-04-15 出版日期: 2020-07-10
ZTFLH:  P237  
基金资助: “十二五”国家科技支撑计划项目(2015BAD23B0202);“双一流”奖补资金(SYL2017XTTD02)
通讯作者: 赵庚星     E-mail:;
作者简介: 董超(1984-),男,山东昌邑人,讲师,主要从事农业信息技术研究。E?mail:
E-mail Alert


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.

Chao Dong,Gengxing Zhao. Influence of Time Series Data Quality on Land Cover Classification Accuracy. Remote Sensing Technology and Application, 2020, 35(3): 558-566.


表1  不同质量时序数据集对照表
图1  研究技术路线图
图2  训练样点分布图


(一季作物、 两季作物)

表2  地类分类对照表
图3  不同处理分类精度
表3  不同处理时序数据集分类精度
图4  不同时间分辨率曲线对比图
图5  时间节点重要性对比图
图6  线性插值处理对比图
图7  HANTS平滑曲线对比
均值5 023.285 022.783 684.123 683.634 841.324 840.8251.9151.992 011.262 010.76
最小值2 666.992 841.891 799.571 848.472 498.692 800.26-1 841.73-1 300.211 011.111 140.41
最大值8 143.528 579.487 881.157 726.977 625.727 308.472 386.401 753.773 059.352 891.24
标准差1 712.041 638.951 840.821 824.661 682.771 657.611 010.20957.94622.89609.75
表4  曲线平滑统计表
训练时间5 s6 s4 s5 s4 s5 s4 s6 s
分类时间36 s37 s36 s37 s34 s35 s33 s34 s
表5  不同处理时序数据集分类时间
图8  分类结果图
1 Yan J W, Liu J Y, Chen B Z,et al. Changes in the Land Surface Energy Budget in Eastern China over the Past Three Decades: Contributions of Land-cover Change and Climate Change[J]. Journal of Climate, 2014, 27(24): 9233-9252.
2 Moon M, Zhang X Y, Henebry G M,et al. Long-term Continuity in Land Surface Phenology Measurements: A Comparative Assessment of the MODIS Land Cover Dynamics and VIIRS Land Surface Phenology Products[J]. Remote Sensing of Environment,2019,226: 74-92.doi:10.1016/j.rse.2019.03.034.
doi: 10.1016/j.rse.2019.03.034
3 Sedlar Z, Alegro Z, Radovic Z,et al. Extreme Land-cover and Biodiversity Change as an Outcome of Land Abandonment on a Mediterranean Island (Eastern Adriatic)[J]. Plant Biosystems, 2018, 152(4): 728-737.
4 Sattler C, Gianuca A T, Schweiger O,et al. Pesticides and Land Cover Heterogeneity Affect Functional Group and Taxonomic Diversity of Arthropods in Rice Agroecosystems[J]. Agriculture Ecosystems & Environment, 2020, 297: 8.doi: 10.1016/j.agee.2020.106927.
doi: 10.1016/j.agee.2020.106927
5 Donnelly S. Land-use Portfolios and the Management of Private Landholdings in South-central Indiana[J]. Regional Environmental Change, 2011, 11(1): 97-109.
6 Li W F, Bai Y, Chen Q W,et al. Discrepant Impacts of Land Use and Land Cover on Urban Heat Islands: A Case Study of Shanghai, China[J]. Ecological Indicators, 2014, 47: 171-178.doi:10.1016/j.ecdind.2014.08.015.
doi: 10.1016/j.ecdind.2014.08.015
7 Vittek M, Brink A, Donnay F,et al. Land Cover Change Monitoring Using Landsat MSS/TM Satellite Image Data over West Africa between 1975 and 1990[J]. Remote Sensing, 2014, 6(1): 658-676.
8 Meneses B M, Reis E, Pereira S,et al. Understanding Driving Forces and Implications Associated with the Land Use and Land Cover Changes in Portugal[J]. Sustainability, 2017, 9(3): 20. doi: 10.3390/su9030351.
doi: 10.3390/su9030351
9 Hansen M C, Potapov P V, Moore R,et al. High-resolution Global Maps of 21st-Century Forest Cover Change[J]. Science, 2013, 342(6160): 850-853.
10 Xu J, Li J S. Case Library Construction Method Using High-resolution Remote Sensing Land Cover Classification Information[J]. Journal of Spatial Science, 2020, 65(1): 173-184.
11 Nakaegawa T. Uncertainty in Land Cover Datasets for Global Land-surface Models Derived from 1 km Global Land Cover Datasets[J]. Hydrological Processes, 2011, 25(17): 2703-2714.
12 Becker-Reshef I, Franch B, Barker B,et al. Prior Season Crop Type Masks for Winter Wheat Yield Forecasting: A US Case Study[J]. Remote Sensing, 2018, 10(10): 20.doi: 10.3390/rs10101659.
doi: 10.3390/rs10101659
13 Holben B. Characteristics of Maximum-value Composite Images from Temporal AVHRR Data[J]. International Journal of Remote Sensing, 1986, 7(11): 1417-1434.
14 Vermote E F, El Saleous N A, Justice C O. Atmospheric Correction of MODIS Data in the Visible to Middle Infrared: First Results[J]. Remote Sensing of Environment, 2002, 83(1): 97-111.
15 Schaaf C B, Gao F, Strahler A H,et al. First Operational BRDF, Albedo Nadir Reflectance Products from MODIS[J]. Remote Sensing of Environment, 2002, 83(1): 135-148.
16 Li Ru,Zhang Xia, Liu Bo, et al. Review on Methods of Remote Sensing Time-series Data Reconstruction[J].Journal of Remote Sensing,2009, 13(2): 335-341.
16 李儒,张霞,刘波,等. 遥感时间序列数据滤波重建算法发展综述[J]. 遥感学报, 2009, 13(2): 335-341.
17 Onojeghuo A O, Blackburn G A, Wang G M,et al. Rice Crop Phenology Mapping at High Spatial and Temporal Resolution Using Downscaled MODIS Time-series[J]. Giscience & Remote Sensing, 2018, 55(5): 659-677.
18 Wang Zhengxing, Ke Linghong, Deng Fangping. Doubling MODIS-NDVI Temporal Resolution :from 16-Day to 8-Day[J]. Remote Sensing Technology and Application, 2011, 26(4): 437-443.
18 王正兴,柯灵红,邓芳萍. MODIS-NDVI时间分辨率加倍算法:从16-Day到8-Day[J]. 遥感技术与应用, 2011, 26(4): 437-443.
19 Olivier J C, Wessels K, Araya S. Detection of Land Cover Change Using an Artificial Neural Network within a Temporal Sliding Windowon Modis Time Series Data[C]// IGARSS 2008 - 2008 IEEE International Geoscience and Remote Sensing Symposium, 2008, doi: 10.1109/IGARSS. 2008.4851314.
doi: 10.1109/IGARSS. 2008.4851314
20 Feng L, Guo S, Zhu L J,et al. Urban Vegetation Phenology Analysis Using High Spatio-temporal NDVI Time Series[J]. Urban Forestry and Urban Greening, 2017, 25: 43-57.doi:10. 1109/JURSE.2015.7120477.
doi: 10. 1109/JURSE.2015.7120477
21 Hmimina G, Dufrene E, Pontailler J Y,et al. Evaluation of the Potential of MODIS Satellite Data to Predict Vegetation Phenology in Different Biomes:An Investigation Using Ground-based NDVI Measurements[J]. Remote Sensing of Environment,2013,132:145-158.doi:10.1016/j.rse.2013.01. 010
doi: 10.1016/j.rse.2013.01. 010
22 Liang S Z, Ma W D, Sui X Y,et al. Extracting the Spatio-temporal Pattern of Cropping Systems from NDVI Time Series Using a Combination of the Spline and HANTS Algorithms: A Case Study for Shandong Province[J]. Canadian Journal of Remote Sensing, 2017, 43(1): 1-15. doi:10.1080/07038992.2017.1252906.
doi: 10.1080/07038992.2017.1252906
23 Ming D P, Zhou T N, Wang M,et al. Land Cover Classification Using Random Forest with Genetic Algorithm-based Parameter Optimization[J]. Journal of Applied Remote Sensing, 2016, 10: 16.doi: 10.1117/1.JRS.10.035021.
doi: 10.1117/1.JRS.10.035021
24 Zuo Y S,Wang W,Hao Y L,et al. Land Cover Classification based on MODIS Images:Taking the Beijing-Tianjin-Hebei Region as an Example[J]. Progress in Geography, 2014, (11): 1556-1565.
25 Li Tao, Wan Guanghua, Zhao Gengxing, et al. Cultivated land in shandong[M]. Beijing: China Agriculture Press. 2018.李涛,万广华,赵庚星,等. 山东耕地[M].北京:中国农业出版社. 2018.
[1] 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.
[2] 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.
[3] 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.
[4] 龙爽,郭正飞,徐粒,周华真,方伟华,许映军. 基于Google Earth Engine的中国植被覆盖度时空变化特征分析[J]. 遥感技术与应用, 2020, 35(2): 326-334.
[5] 唐廷元,付波霖,何素云,娄佩卿,闭璐. 基于GF-1和Sentinel-1A的漓江流域典型地物信息提取[J]. 遥感技术与应用, 2020, 35(2): 448-457.
[6] 郑贵洲,熊良超,廖艳雯,王红平. 利用MODIS数据反演南海南部海表温度及时空变化分析[J]. 遥感技术与应用, 2020, 35(1): 132-140.
[7] 赵春亮,许文波,范锦龙. FY-3C中分辨率成像光谱仪数据的窄波段地表反照率验证研究[J]. 遥感技术与应用, 2020, 35(1): 153-162.
[8] 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.
[9] 徐卫星,薛华柱,靳华安,李爱农. 融合遥感先验信息的叶面积指数反演[J]. 遥感技术与应用, 2019, 34(6): 1235-1244.
[10] 刘培,余志远,马威,韩瑞梅,陈正超,王涵,杨磊库. 基于地形信息的Landsat与Radarsat-2遥感数据协同分类研究[J]. 遥感技术与应用, 2019, 34(6): 1269-1275.
[11] 李哲,张沁雨,彭道黎. 基于高分二号遥感影像的树种分类方法[J]. 遥感技术与应用, 2019, 34(5): 970-982.
[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] 冯艾琳,武晋雯,孟莹,姜鹏,董巍,张璇,方缘,刘斌. 基于MODIS GPP数据产品的辽宁省碳源/汇空间格局分布研究[J]. 遥感技术与应用, 2019, 34(4): 857-864.
[14] 冯婵莹, 汪子豪, 郑成洋. 基于分层线性回归的MODIS反照率产品降尺度方法研究[J]. 遥感技术与应用, 2019, 34(3): 602-611.
[15] 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.