黑河流域时间序列植被指数曲线的平滑去噪对比研究
Study on the Smooth Denoising Contrast of the Vegetation Index Curve of the Time Series of the Heihe River Basin
DOI: 10.12677/AG.2018.81004, PDF,   
作者: 和 萍*, 高书鹏：云南师范大学旅游与地理科学学院，云南 昆明
关键词: 时间序列数据；MODIS NDVI；去噪；TIMESAT3.1；黑河流域；Time Series Data； MODIS NDVI； Denoising； TIMESAT3.1； Heihe Basin
摘要: 长时间序列遥感数据集被广泛应用于全球或区域环境变化、植被动态变化、土地覆盖变化和植物生物物理参数反演等研究，受云、气溶胶、太阳高度角等因素的影响使数据存在很多的噪声。影响了数据分析和应用的效果。本文以MODIS-NDVI数据为基础，黑河流域为研究区，在MATLAB编程环境下，利用时间序列拟合软件TIMESAT3.1，使用Savitzky-Golay (S-G)滤波、非对称高斯函数(AG)和双逻辑曲线拟合(DL)方法对黑河流域2012年NDVI时间序列数据进行了重构。结果表明，TIMESAT3.1软件能有效的对NDVI时间序列进行平滑去噪，三种方法在时间序列曲线上均取得了较好的效果。但是不对称高斯函数和双逻辑曲线拟合结果在植被的生长季起始阶段均存在“翘起”的过度拟合问题。相对而言，S-G滤波拟合效果较好，NDVI曲线在保持原有基本形状的基础上更加有效的揭示所蕴含的物候周期性变化规律，该数据源对黑河流域的植被检测提供了良好的基础。
Abstract: Long time series remote sensing data set is widely applied to global or regional environmental change, vegetation dynamic change, land cover change and plant biophysical parameters inversion. It is affected by clouds, aerosols, solar elevation and other factors, which make the data has a lot of noise. This paper is based on the MODIS-NDVI data, the Heihe River Basin, in the MATLAB programming environment, using time series fitting software TIMESAT3.1, using Savitzky-Golay (S-G) filter, asymmetric Gauss function (AG) and dual logic curve fitting (DL) method for the re-construction of NDVI time series data of Heihe River Basin in 2012. The results show that the TIMESAT3.1 software can effectively denoise the NDVI time series, and the three methods have achieved good results on the time series curve. However, the unsymmetrical Gauss function and the double logical curve fitting results have the overfitting problem of "warping" in the beginning of the growing season of vegetation. In contrast, the S-G filter has a good fitting effect. The data source provides a good basis for the vegetation detection in the Heihe River Basin.
文章引用：和萍, 高书鹏. 黑河流域时间序列植被指数曲线的平滑去噪对比研究[J]. 地球科学前沿, 2018, 8(1): 32-41. https://doi.org/10.12677/AG.2018.81004

参考文献

[1] 黄耀欢, 王建华, 江东, 等. 利用S-G滤波进行MODIS-EVI时间序列数据重构[J]. 武汉大学学报(信息科学版), 2009(12): 1440-1443.
[2] 赵英时. 遥感应用分析原理与方法[M]. 科学出版社, 2003.
[3] 岑奕, 张良培, 村松加奈子. 纪伊半岛地区植被净初级生产力的遥感应用研究[J]. 武汉大学学报(信息科学版), 2008, 33(12): 1221-1224.
[4] Moody, A. and Johnson, D.M. (2001) Land-Surface Phenologies from AVHRR Using the Discrete Fourier Transform. Remote Sensing of Environment, 75, 305-323. [Google Scholar] [CrossRef] 
[5] Feng, Z., Wu, B., Liu, C., et al. (2004) Methods of Monitoring Crop Phonological Stages Using Time Series of Vegetation Indicator. Transactions of the Chinese Society of Agricultural Engineering, 20, 155-159.
[6] Juan, G.U., Xin, L.I. and Huang, C.L. (2006) Research on the Reconstructing of Time-Series NDVI Data. Remote Sensing Technology & Application, 21, 391-395.
[7] Bian, J., Ainong, L.I., Song, M., et al. (2010) Reconstruction of NDVI Time-Series Datasets of MODIS Based on Savitzky-Golay Filter. Journal of Remote Sensing, 14, 725-741.
[8] 边金虎, 李爱农, 宋孟强, 等. MODIS植被指数时间序列Savitzky-Golay滤波算法重构[J]. 遥感学报, 2010, 14(4): 725-741.
[9] 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.
[10] Ma, M. and Veroustraete, F. (2006) Reconstructing Pathfinder AVHRR Land NDVI Time-Series Data for the Northwest of China. Advances in Space Research, 37, 835-840. [Google Scholar] [CrossRef] 
[11] 孙华生, 徐爱功, 林卉, 等. 基于不同算法的时间序列植被指数去噪效果分析[J]. 江苏农业科学, 2012, 40(5): 375-379.
[12] Bradley, B.A., Jacob, R.W., Hermance, J.F., et al. (2007) A Curve Fitting Procedure to Derive Inter-Annual Phenologies from Time Series of Noisy Satellite NDVI Data. Remote Sensing of Environment, 106, 137-145. [Google Scholar] [CrossRef] 
[13] Chen, J., Jönsson, P., Tamura, M., et al. (2004) A Simple Method for Recon-structing a High-Quality NDVI Time-Series Data Set Based on the Savitzky-Golay Filter. Remote Sensing of Environment, 91, 332-344. [Google Scholar] [CrossRef] 
[14] 吴文斌, 杨鹏, 唐华俊, 等. 两种NDVI时间序列数据拟合方法比较[J]. 农业工程学报, 2009, 25(11): 183-188.
[15] Taddei, R. (1997) Maximum Value Interpolated (MVI): A Maximum Value Composite Me-thod Improvement in Vegetation Index Profiles Analysis. International Journal of Remote Sensing, 18, 2365-2370. [Google Scholar] [CrossRef] 
[16] 卫炜, 吴文斌, 李正国, 等. 时间序列植被指数重构方法比对研究[J]. 中国农业资源与区划, 2014, 35(1): 34-43.
[17] 耿丽英, 马明国. 长时间序列NDVI数据重建方法比较研究进展[J]. 遥感技术与应用, 2014, 29(2): 362-368.
[18] Viovy, N., Arino, O. and Belward, A.S. (1992) The Best Index Slope Extraction (BISE): A Method for Reducing Noise in NDVI Time-Series. International Journal of Remote Sensing, 13, 1585-1590. [Google Scholar] [CrossRef] 
[19] Lovell, J.L. and Graetz, R.D. (2001) Filtering Pathfinder AVHRR Land NDVI Data for Australia. International Journal of Remote Sensing, 22, 2649-2654. [Google Scholar] [CrossRef] 
[20] Wang, D., Jiang, X.G., Tang, L.L. and Xi, X. (2005) The Application of Time-Series Fourier Analysis to Reconstructing Cloud-Free NDVI Images. Remote Sensing for Land & Resources, 16, 29-32.
[21] Roerink, G.J., Menenti, M. and Verhoef, W. (2000) Reconstructing Cloudfree NDVI Composites using Fourier Analysis of Time Series. International Journal of Remote Sensing, 21, 1911-1917. [Google Scholar] [CrossRef] 
[22] 宋春桥, 柯灵红, 游松财, 等. 基于TIMESAT的3种时序NDVI拟合方法比较研究——以藏北草地为例[J]. 遥感技术与应用, 2011(2): 147-155.
[23] 曹云锋, 王正兴, 邓芳萍. 3种滤波算法对NDVI高质量数据保真性研究[J]. 遥感技术与应用, 2010, 25(1): 118-125.
[24] Savitzky, A. and Golay, M.J.E. (1964) Smoothing and Differentiation of Data by Simplified Least Squares Procedures. Analytical Chemistry, 36, 1627-1639. [Google Scholar] [CrossRef] 
[25] Luo, J., Ying, K. and Bai, J. (2005) Savitzky-Golay Smoothing and Differentiation Filter for Even Number Data. Signal Processing, 85, 1429-1434. [Google Scholar] [CrossRef] 
[26] Zhu, M.Z. and Bao, X. (2008) Application of Outlier Detection and Savitzky-Golay Filter in Handwritten Systems. Mechanical & Electrical Engineering Magazine, 8, 5-7.
[27] Jonsson, P. and Eklundh, L. (2002) Seasonality Extraction by Function Fitting to Time-Series of Satellite Sensor Data. IEEE Transactions on Geoscience & Remote Sensing, 40, 1824-1832. [Google Scholar] [CrossRef] 
[28] Beck, P.S.A., Atzberger, C., Høgda, K.A., et al. (2006) Improved Monitoring of Vegetation Dynamics at Very High Latitudes: A New Method using MODIS NDVI. Remote Sensing of Environment, 99, 321-334. [Google Scholar] [CrossRef] 
[29] Jönsson, P. and Eklundh, L. (2004) TIMESAT—A Program for Analyzing Time-Series of Satellite Sensor Data. Computers & Geosciences, 30, 833-845. [Google Scholar] [CrossRef] 
[30] Cao, Y.F., Wang, Z.X. and Den, F.P. (2010) Fidelity Performance of Three Filters for High Quality NDVI Time-Series Analysis. Remote Sensing Technology & Application, 25, 118-125.
[31] Song, C., You, S., Ke, L., et al. (2011) Analysis on Three NDVI Time-Series Reconstruction Methods and Their Applications in North Tibet. Journal of Geo-Information Science, 13, 133-143. [Google Scholar] [CrossRef]