Hyperspectral Retrieval Model of Soil Organic Matter Content Based on Particle Swarm Optimization-Support Vector Machines
-
摘要: 为了监测复垦矿区土壤的有机质含量, 综合利用光谱分析、统计学习理论与方法以及智能优化理论与方法, 研究了矿区复垦土壤有机质含量与土壤光谱之间的关系, 在此基础上建立了土壤有机质含量高光谱反演模型, 实现土壤有机质含量定量检测.首先对原始土壤光谱数据进行预处理, 然后进行相关性分析, 提取450 nm、500 nm、650 nm、770 nm、1 460 nm和2 140 nm作为特征波段, 最后利用多元线性回归(multiple linear regression, MLR)、偏最小乘回归(partial least squares regression, PLSR)和粒子群优化支持向量机回归(particle swarm optimization support vector machine regression, PSO-SVM)方法建立了土壤有机质含量的高光谱定量反演模型, 并对模型进行验证.3种模型的验证结果如下: MLR、PLSR和PSO-SVM模型的R2分别为0.79、0.83和0.85, RMSE分别为5.26、4.93和4.76.实验结果表明, 无论从模型的稳定性还是预测能力上, PSO-SVM都要优于其他两个模型.
-
关键词:
- 土壤有机质 /
- 高光谱 /
- 遥感 /
- 粒子群优化支持向量机 /
- 粒子群算法
Abstract: To monitor the soil organic matter in the reclamation area of coal mines, the relationship between soil organic matter content and soil spectra in the reclamation area of coal mines was studied, and a quantitative retrieval model was established and validated in order to implement the organic matter content detection in this paper. After the preprocessing of the original spectral, the correlation of the organic matter content and reflectance spectra was analyzed, and 450 nm, 500 nm, 650 nm, 770 nm, 1 460 nm and 2 140 nm wavelength were extracted as feature bands. Using the multiple linear regression (MLR), partial least squares regression (PLSR) and particle swarm optimization support vector machine regression (PSO-SVM) methods, the hyperspectral quantitative retrieval models for soil organic matter content were built. The results show the coefficient of determination (R2) of MLR, PLSR and PSO-SVM were 0.79, 0.83 and 0.85 respectively, and the root mean square error of prediction (RMSEP) were 5.26, 4.93 and 4.76 respectively. The results demonstrate that the stability and predictive ability of PSO-SVM model are better than those of the MLR and PLSR model.-
Key words:
- soil organic matter /
- hyperspectral /
- remote sensing /
- PSO-SVM /
- particle swarm optimization algorithm
-
图 3 土壤有机质含量与不同变换形式光谱的相关系数随波长变化
a.与平滑光谱；b.与一阶微分光谱；c.与标准正态变量变换光谱；d.与连续统去除光谱
Fig. 3. The distribution of the correlation coefficients between the SOM concentrations and the different transformed reflectance spectra
图 4 不同模型有机质实测值与模型预测值之间的比较
Fig. 4. A comparison between the measured values and predicted values of the different models
表 1 土壤有机质含量检测统计结果
Table 1. The statistical results of the SOM
最大值(g/kg) 最小值(g/kg) 平均值(g/kg) 标准差(g/kg) 有机质 53.00 8.50 29.53 12.64 
下载: 导出CSV
表 2 土壤有机质含量定量反演模型的特征光谱
Table 2. The characteristic spectra of the quantitative inversion model of SOM
变量名 波段(nm) 相关系数 SNV(R) 450 0.87 SNV(R) 500 0.90 CR(R) 650 0.83 CR(R) 770 0.87 SNV(R) 1 460 -0.81 CR(R) 2 140 0.87
下载: 导出CSV
表 3 土壤有机质含量预反演模型的精度统计结果
Table 3. The results of the quantitative inversion models of SOM
模型 R2 RMSE MLR 0.79 5.26 PLSR 0.83 4.93 PSO-SVM 0.85 4.76
下载: 导出CSV
-
[1] Chu, X.L., Yuan, H.F., Lu, W.Z., 2004. Progress and Application of Spectral Data Pretreatment and Wavelength Selection Methods in NIR Analytical Technique. Progress in Chemistry, 16(4): 528-542(in Chinese with English abstract). http://www.researchgate.net/publication/279620861_Progress_and_application_of_spectral_data_pretreatment_and_wavelength_selection_method_in_NIR_analytical_technique [2] Demattê, J.A.M., Sousa, A.A., Alves, M.C., et al., 2006. Determining Soil Water Status and Other Soil Characteristics by Spectral Proximal Sensing. Geoderma, 135: 179-195. doi: 10.1016/j.geoderma.2005.12.002 [3] Hair, J.F., Sarstedt, M., Ringle, C.M., et al., 2012. An Assessment of the Use of Partial Least Squares Structural Equation Modeling in Marketing Research. Journal of the Academy of Marketing Science, 40(3): 414-433. doi: 10.1007/s11747-011-0261-6 [4] He, J.L., Jiang, J.J., Zhou, S.L., et al., 2007. The Hyperspectral Characteristics and Retrieval of Soil Organic Matter Content. Scientia Agricultura Sinica, 40(3): 638-643(in Chinese with English abstract). http://agris.fao.org/agris-search/search.do?recordID=CN2007000678 [5] Hummel, J.W., Sudduth, K.A., Hollinger, S.E., 2001. Soil Moisture and Organic Matter Prediction of Surface and Subsurface Soils Using an Nir Soil Sensor. Computers and Electronics in Agriculture, 32(2): 149-165. doi: 10.1016/S0168-1699(01)00163-6 [6] Kennedy, J., 2010. Particle Swarm Optimization. Springer, New York, 760-766. [7] Liu, H.J., Zhang, B., Zhao, J., et al., 2007. Spectral Models for Prediction of Organic Matter in Black Soil. Acta Petrologica Sinica, 44(1): 27-32(in Chinese with English abstract). http://pedologica.issas.ac.cn/trxben/ch/reader/view_abstract.aspx?file_no=20070105&flag=1 [8] Liu, H.J., Zhang, X.L., Zheng, S.F., et al., 2010. Black Soil Organic Matter Predicting Model Based on Field Hyperspectral Reflectance. Spectroscopy and Spectral Analysis, 30(12): 3355-3358(in Chinese with English abstract). http://europepmc.org/abstract/MED/21322239 [9] Liu, W.D., Baret, F., Gu, X.F., et al., 2002. Relating Soil Surface Moisture to Reflectance. Remote Sensing of Environment, 81(2-3): 238-246. doi: 10.1016/S0034-4257(01)00347-9 [10] Lu, Y.L., Bai, Y.L., Yang, L.P., et al., 2008. Application of Hyperspectral Data for Soil Organic Matter Estimation Based on Principle Components Regerssion Analysis. Plant Nutrition and Fertilizer Science, 14(6): 1076-1082(in Chinese with English abstract). http://www.en.cnki.com.cn/Article_en/CJFDTOTAL-ZWYF200806010.htm [11] Luan, F.M., Zhang, X.L., Xiong, H.G., et al., 2013. Comparative Analysis of Soil Organic Matter Content Based on Different Hyperspectral Inversion Models. Spectroscopy and Spectral Analysis, 33(1): 196-200(in Chinese with English abstract). http://www.ncbi.nlm.nih.gov/pubmed/23586255 [12] Ren, H.Y., Zhuang, D.F., Singh, A., et al., 2009. Estimation of As and Cu Contamination in Agricultural Soils around a Mining Area by Reflectance Spectroscopy: A Case Study. Pedosphere, 19(6): 719-726. doi: 10.1016/S1002-0160(09)60167-3 [13] Sha, J.M., Chen, P.C., Lin, C.S., 2003. Characteristics Analysis of Soil Spectrum Response Resulted from Organic Material. Research of Soil and Water Conservation, 10(2): 21-24(in Chinese with English abstract). http://en.cnki.com.cn/Article_en/CJFDTOTAL-STBY200302005.htm [14] Shen, R.P., Ding, G.X., Wei, G.S., et al., 2009. Retrieval of Soil Organic Matter Content from Hyper-Spectrum Based on Ann. Acta Petrologica Sinica, 46(3): 391-397(in Chinese with English abstract). http://www.cabdirect.org/abstracts/20093255951.html [15] Smola, A.J., Schölkopf, B., 2004. A Tutorial on Support Vector Regression. Statistics and Computing, 14(3): 199-222. doi: 10.1023/B:STCO.0000035301.49549.88 [16] Svensson, O., Kourti, T., Macgregor, J.F., 2002. An Investigation of Orthogonal Signal Correction Algorithms and Their Characteristics. Journal of Chemometrics, 16(4): 176-188. doi: 10.1002/cem.700 [17] Wang, J., He, T., Li, Y.H., 2005. Studying on Extraction Methods for Land Quality Information Based on Hyperspectral Data. Journal of Remote Sensing, 9(4): 438-445(in Chinese with English abstract). [18] Wold, H., 1985. Partial Least Squares. Encyclopedia of Statistical Sciences. John Wiley & Sons. Inc. USA, 9. doi: 10.1002/0471667196.ess1914.pub2 [19] Xu, Y.M., Zhong, L.Q., Wang, L., et al., 2006. Model for Estimating Soil Nutrient Elements Based on High Resolution Reflectance Spectra. Acta Petrologica Sinica, 43(5): 709-716(in Chinese with English abstract). http://en.cnki.com.cn/Article_en/CJFDTOTAL-TRXB200605000.htm [20] Zheng, L.H., Li, M.Z., An, X.F., et al., 2010. Forecasting Soil Parameters Based on NIR and SVM. Transactions of the CSAE, 26(Suppl. 2): 81-87(in Chinese with English abstract). http://www.cqvip.com/QK/90712X/2010S2/3000101830.html [21] Zheng, L.H., Li, M.Z., Pan, L., et al., 2008. Estimation of Soil Organic Matter and Soil Total Nitrogen Based on NIR Spectroscopy and BP Neural Network. Spectroscopy and Spectral Analysis, 28(5): 1160-1164(in Chinese with English abstract). http://en.cnki.com.cn/Article_en/CJFDTOTAL-GUAN200805050.htm [22] Zuo, L., Hou, L.G., Zhang, W., et al., 2010. Analog Circuit Fault Diagnosis Based on Partical Swarm Optimization Support Vector Machine. Systems Engineering and Electronics, 32(7): 1553-1556(in Chinese with English abstract). http://www.researchgate.net/publication/292252220_Analog_circuit_fault_diagnosis_based_on_particle_swarm_optimization_support_vector_machine [23] 褚小立, 袁洪福, 陆婉珍, 2004. 近红外分析中光谱预处理及波长选择方法进展与应用. 化学进展, 16(4): 528-542. doi: 10.3321/j.issn:1005-281X.2004.04.008 [24] 贺军亮, 蒋建军, 周生路, 等, 2007. 土壤有机质含量的高光谱特性及其反演. 中国农业科学, 40(3): 638-643. doi: 10.3321/j.issn:0578-1752.2007.03.030 [25] 刘焕军, 张柏, 赵军, 等, 2007. 黑土有机质含量高光谱模型研究. 土壤学报, 44(1): 27-32. doi: 10.3321/j.issn:0564-3929.2007.01.005 [26] 刘焕军, 张新乐, 郑树峰, 等, 2010. 黑土有机质含量野外高光谱预测模型. 光谱学与光谱分析, 30(12): 3355-3358. doi: 10.3964/j.issn.1000-0593(2010)12-3355-04 [27] 卢艳丽, 白由路, 杨俐苹, 等, 2008. 基于主成分回归分析的土壤有机质高光谱预测与模型验证. 植物营养与肥料学报, 14(6): 1076-1082. doi: 10.3321/j.issn:1008-505X.2008.06.008 [28] 栾福明, 张小雷, 熊黑钢, 等, 2013. 基于不同模型的土壤有机质含量高光谱反演比较分析. 光谱学与光谱分析, 33(1): 196-200. doi: 10.3964/j.issn.1000-0593(2013)01-0196-05 [29] 沙晋明, 陈鹏程, 陈松林, 2003. 土壤有机质光谱响应特性研究. 水土保持研究, 10(2): 21-24. doi: 10.3969/j.issn.1005-3409.2003.02.006 [30] 沈润平, 丁国香, 魏国栓, 等, 2009. 基于人工神经网络的土壤有机质含量高光谱反演. 土壤学报, 46(3): 391-397. doi: 10.3321/j.issn:0564-3929.2009.03.003 [31] 王静, 何挺, 李玉环, 2005. 基于高光谱遥感技术的土地质量信息挖掘研究. 遥感学报, 9(4): 438-445. https://www.cnki.com.cn/Article/CJFDTOTAL-YGXB200504014.htm [32] 徐永明, 蔺启忠, 王璐, 等, 2006. 基于高分辨率反射光谱的土壤营养元素估算模型. 土壤学报, 43(5): 709-716. doi: 10.3321/j.issn:0564-3929.2006.05.001 [33] 郑立华, 李民赞, 安晓飞, 等, 2010. 基于近红外光谱和支持向量机的土壤参数预测. 农业工程学报, 26(增刊2): 81-87. https://www.cnki.com.cn/Article/CJFDTOTAL-NYGU2010S2017.htm [34] 郑立华, 李民赞, 潘娈, 等, 2008. 基于近红外光谱技术的土壤参数BP神经网络预测. 光谱学与光谱分析, 28(5): 1160-1164. doi: 10.3964/j.issn.1000-0593.2008.05.047 [35] 左磊, 侯立刚, 张旺, 等, 2010. 基于粒子群支持向量机的模拟电路故障诊断. 系统工程与电子技术, 32(7): 1553-1556. doi: 10.3969/j.issn.1001-506X.2010.07.047 -
点击查看大图
图(4) / 表(3)
计量
- 文章访问数: 3222
- HTML全文浏览量: 189
- PDF下载量: 574
- 被引次数: 0
