The Estimation Model of Rice Leaf Area Index Using Hyperspectral Data Based on Support Vector Machine
YANG Xiao-hua1,3,HUANG Jing-feng1,WANG Xiu-zhen2,WANG Fu-min1*
1.Institute of Remote Sensing & Information Application, Zhejiang University, Hangzhou 310029, China 2.Zhejiang Meteorological Institute, Hangzhou 310004, China 3.Meteorological and Hydrographic Department of General Staff Headquarters, Beijing 100081, China
Abstract：In order to compare the prediction powers between the best statistical model and SVM technique using each VI for rice LAI, the VIs are as independent variables in statistical models and are as net inputs in SVM, and the rice LAI are as dependent variables in statistical models and are as net outputs in SVM.Hyperspectral reflectance (350 to 2 500 nm) data were recorded in two experiments involving four replicates of two rice cultivars (“Xiushui 110” and “Xieyou 9308”), three nitrogen levels (0, 120, 240 kg·ha-1 N), and with a plant density of 45 plants·m-2.The first experiment was seeded on 30 May 2004 and the second experiment on 15 June 2004.Both sets of seedlings were transplanted to the field one month later.Hyperspectral reflectance was ground-based and measured using Analytical Spectral DevicesTM and 1 meter above the rice canopy.The solar angle compared to nadir was for all measurements less than 45° and no disturbing clouds were observed.Hyperspectral reflectance was transformed to ten different vegetation indices including RVI, NDVI, NDVIgreen, SAVI, OSAVI, MSAVI, MCACI, TCARI/OSAVI, RDVI and RVI2, according to the width of TM bands of Ladsat-5.Different statistical models including linearity model, exponent model, power model and logarithm model, were analyzed using all samples’ LAI and vegetation indices.Three good relationships including exponent relationship of NDVIgreen, power relationship of TCARI/OSAVI and power relationship of RV12 were selected based on the R2 of models.These three relationships were used to predict the LAI of rice through SVM models with different kernel functions including an analysis of variance kernel (ANOVA), a polynomial kernel (POLY) and a radial basic function kernel (RBF), and corresponding statistical models.The results show that all SVM models have lower RMSE values and higher estimation precision than corresponding statistical models；SVM with POLY kernel function using TCARI/OSAVI has the highest estimation precision for rice LAI compared to other models, and it’s RMSE value is lower than corresponding statistical model by 11 percent points.Therefore, SVM has a high accuracy for learning and a good robustness for estimation of LAI of rice using hyperspectral data.Consequently, SVM provides a useful explorative tool for improvement of the relationships between VIs and rice LAI.
Key words：Support Vector Machine;Hyperspectral;Leaf Area Index
杨晓华1,3,黄敬峰1,王秀珍2,王福民1*. 基于支持向量机的水稻叶面积指数高光谱估算模型研究[J]. 光谱学与光谱分析, 2008, 28(08): 1837-1841.
YANG Xiao-hua1,3,HUANG Jing-feng1,WANG Xiu-zhen2,WANG Fu-min1*. The Estimation Model of Rice Leaf Area Index Using Hyperspectral Data Based on Support Vector Machine. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2008, 28(08): 1837-1841.
[1] Jamieson P D, Porter J R, Goudrian J, et al.Field Crops Research, 1998, 55：23. [2] Asseng S, Keulen H, Stol W.European Journal of Agronomy, 2000, 12：37. [3] Cheng Q, Huang J, Wang X, et al.Journal of Zhejiang University·A Science, 2003, 4(6)：727. [4] Serrano L, Filella I, Penuelas J.Crop Science, 2000, 40：723. [5] Thenkabail P S, Smith R B, de Pauw E.Remote Sensing of Environment, 2000, 71：158. [6] Tang Y, Wang R, Huang J.Pedosphere, 2004, 14(4)：467. [7] Cheng Q.Pedosphere, 2006, 16(3)：362. [8] Gitelson A A, Kaufman Y J, Merzlyak M N.Remote Sensing of Environment, 1996, 58：289. [9] Haboudane D, Miller J R, Tremblay N, et al.Remote Sensing of Environment, 2002, 81：416. [10] Broge N H, Leblanc E.Remote Sensing of Environment, 2000, 76：156. [11] Broge N H, Mortensen J V.Remote Sensing of Environment, 2002, 81：45. [12] Haboudane D, Miller J R, Tremblay N, et al.Remote Sensing of Environment, 2002, 81：416. [13] Keiner L E, Yan X H.Remote Sensing of Environment, 1998, 66：153. [14] Vapnik V N.The Nature of Statistical Learning Theory.NY：Springer-Verlag, 1995. [15] Drucker H, Wu D, Vapnik V.IEEE Trans.Neural Networks, 1999, 10(5)：1048. [16] MA Qun, HAO Gui-qi, QIAO Yan-jiang, et al(马 群, 郝贵奇, 乔延江, 等).Spectroscopy and Spectral Analysis(光谱学与光谱分析), 2006, 26(10)：1842. [17] ZHANG Lu-da, SU Shi-guang, WANG Lai-sheng, et al(张录达, 苏时光, 王来生, 等).Spectroscopy and Spectral Analysis(光谱学与光谱分析), 2005, 25(1)：33. [18] ZHANG Lu-da, JING Ze-huan, SHEN Xiao-nan, et al(张录达, 金泽宸, 沈晓南, 等).Spectroscopy and Spectral Analysis(光谱学与光谱分析), 2005, 25(9)：1400. [19] Cao L.Neurocomputing, 2003, 51：321. [20] ZHANG Xue-gong(张学工).Acta Automatica Sinica(自动化学报)，2000, 26(1)：32.
