张小鸣,汤宁

• 1:常州大学(武进校区)信息科学与工程学院

摘要(Abstract)：

近红外光谱技术是检测土壤信息的有效工具,为了提高预测模型的准确度和建模效率,需要对波长进行优选。提出SiPLS-GA-SPA特征波长提取方法,即协同区间偏最小二乘算法(SiPLS)、遗传算法(GA)和连续投影算法(SPA)对土壤有机质特征波长进行梯度提取,最终从1 050个波长中提取9个土壤有机质的特征波长。利用偏最小二乘回归(PLSR)和支持向量机回归(SVMR)建立6种基于特征波长的土壤有机质含量预测模型。结果表明:SiPLS-GA-SPA-SVMR模型的预测结果为RMSEP=1.15,R2=0.91,优于其他模型;SiPLS-GA-SPA特征波长提取方法能够简化预测模型,提高模型预测精度,为开发便携式近红外光谱土壤养分检测仪提供理论基础。

关键词(KeyWords)： 近红外光谱;特征波长;协同区间偏最小二乘;遗传算法;连续投影算法;支持向量机回归

Abstract:

Keywords:

基金项目(Foundation):

作者(Author): 张小鸣,汤宁

DOI: 10.16652/j.issn.1004-373x.2018.22.031

参考文献(References)：

• [1]宋海燕.土壤近红外光谱检测[M].北京:化学工业出版社,2013.SONG Haiyan. The soil′s near infrared spectroscopy detection[M]. Beijing:Chemical Industry Press,2013.
• [2]刘燕德,熊松盛,刘德力.近红外光谱技术在土壤成分检测中的研究进展[J].光谱学与光谱分析,2014,34(10):2639-2644.LIU Yande,XIONG Songsheng,LIU Deli. Application of near infrared reflectance spectroscopy technique(NIRS)to soil attributes research[J]. Spectroscopy and spectral analysis,2014,34(10):2639-2644.
• [3] RINNAN R,RINNAN A. Application of near infrared reflectance(NIR)and fluorescence spectroscopy to analysis of microbiological and chemical properties of arctic soil[J]. Soil biology&biochemistry,2007,39(7):1664-1673.
• [4]杨峰,张勇,谌俊旭,等.高光谱数据预处理对大豆叶绿素密度反演的作用[J].遥感信息,2017,32(4):64-69.YANG Feng,ZHANG Yong,CHEN Junxu,et al. Effects of hyperspectral data pretreatment on model inversion of soybean chlorophyll density[J]. Remote sensing information,2017,32(4):64-69.
• [5]陈奕云,齐天赐,黄颖菁,等.土壤有机质含量可见-近红外光谱反演模型校正集优选方法[J].农业工程学报,2017,33(6):107-114.CHEN Yiyun,QI Tianci,HUANG Yingjing,et al. Optimization method of calibration dataset for VIS-NIR spectral inver-sion model of soil organic matter content[J]. Transactions ofthe Chinese Society of Agricultural Engineering, 2017, 33(6):107-114.
• [6] GALV?O R K H,ARAUJO M C U,JOSéG E,et al. A meth-od for calibration and validation subset partitioning[J]. Talan-ta,2005,67(4):736-740.
• [7]王瑛瑛.土壤有机质近红外光谱分析及相关软件开发[D].合肥:中国科学技术大学,2014.WANG Yingying. The near-infrared spectroscopy analysis of organic matter and related software development[D]. Hefei:University of Science and Technology of China,2014.
• [8] YANG M,CHEN Q,KUTSANEDZIE F Y H,et al. Portable spectroscopy system determination of acid value in peanut oil based on variables selection algorithms[J]. Measurement,2017,103:179-185.
• [9]杨海清,祝旻.基于可见-近红外光谱特征波长选择的土壤有机质快速检测研究[J].红外,2015,36(2):42-48.YANG Haiqing,ZHU Min. Study of rapid detection of soil organic matter based on characteristic wavelength selection of visible-near infrared spectra[J]. Infrared,2015,36(2):42-48.
• [10]宾俊,范伟,周冀衡,等.智能优化算法应用于近红外光谱波长选择的比较研究[J].光谱学与光谱分析,2017,37(1):95-102.BIN Jun,FAN Wei,ZHOU Jiheng,et al. Application of intelligent optimization algorithms to wavelength selection of near-infrared spectroscopy[J]. Spectroscopy and spectral analysis,2017,37(1):95-102.
• [11]章海亮,罗微,刘雪梅,等.应用遗传算法结合连续投影算法近红外光谱检测土壤有机质研究[J].光谱学与光谱分析,2017,37(2):584-587.ZHANG Hailiang,LUO Wei,LIU Xuemei,et al. Measurement of soil organic matter with near infrared spectroscopy combined with genetic algorithm and successive projection algorithm[J]. Spectroscopy and spectral analysis, 2017, 37(2):584-587.
• [12] MORELLOS A,PANTAZI X E,MOSHOU D,et al. Machine learning based prediction of soil total nitrogen,organic carbon and moisture content by using VIS-NIR spectroscopy[J].Biosystems engineering,2016,152:104-116.
• [13] ROSSEL R A V,BEHRENS T,GUERRERO C,et al. Using data mining to model and interpret soil diffuse reflectance spectra[J]. Geoderma,2010,158(1):46-54.