长波近红外光谱在土壤总氮分析与稳定性中的应用  

Long-Wave NIR Spectroscopy Applied to Analysis of Total Nitrogen in Soil and Stability

作　　者：陈定星[1] 潘涛[1,2] 陈洁梅[2,3] 姚立军[1,2] 

机构地区：[1]光电信息与传感技术广东普通高校重点实验室(暨南大学),广东广州510632 [2]土壤与农业可持续发展国家重点实验室(中国科学院南京土壤研究所),江苏南京210008 [3]暨南大学生物工程学系,广东广州510632

出　　处：《安徽农业科学》2013年第2期567-569,共3页Journal of Anhui Agricultural Sciences

基　　金：国家自然科学基金(No.61078040);中央高校基本科研业务费专项资金(No.21611513);土壤与农业可持续发展国家重点实验室(中国科学院南京土壤研究所)开放课题基金(No.0812201201)

摘　　要：[目的]通过定标集、预测集、检验集的建模过程,采用偏最小二乘(PLS)方法结合波段选择建立土壤总氮快速分析的近红外(NIR)光谱模型。[方法]为了避免模型评价失真,基于随机性、相似性和稳定性,提出一种严谨的建模体系。将全谱扫描区(400～2 498nm)分成可见区(400～780 nm)、短波近红外区(780～1 100 nm)和长波近红外区(1 100～2 498 nm)。[结果]经过比较、检验,结果表明长波近红外达到了最好的模型效果和稳定性,最优PLS因子数为8,检验的预测均方根误差(V-SEP)和预测相关系数(V-RP)分别为0.118 g/kg和0.857,得到客观、稳定的预测模型。[Objective] Through modeling process of the calibration set,the prediction set and validation set,the near-infrared（NIR） spectroscopy model for rapid analysis of total nitrogen in soil was established by using partial least squares（PLS） method combined with waveband selection.[Method] In order to avoid the evaluation distortion for modeling,a rigorous modeling system was proposed based on randomness,similarity and stability.The overall scanning region（400-2 498 nm） was separated into the visible region（400-780 nm）,the short-wave NIR region（780-1 100 nm）,and the long-wave NIR region（1 100-2 498 nm）.[Result] By comparison and validation,the results indicated that the long-wave NIR region had the best prediction accuracy and stability,and the optimal PLS factor was 8,the validation root mean square error of prediction（V-SEP） and the validation correlation coefficient of prediction（V-RP） were 0.118 g/kg and 0.857,respectively,anobjective and stable prediction model was obtained.

关 键 词：土壤 总氮 长波近红外光谱 偏最小二乘 稳定性 

分 类 号：S153.61[农业科学—土壤学]