土壤速效磷含量近红外光谱田间快速测定方法  

Detection of Soil Available P Content Based on Near Infrared Spectra

作　　者：何东健[1] 陈煦[1] 任嘉琛 王梅嘉 

机构地区：[1]西北农林科技大学机械与电子工程学院,陕西杨凌712100 [2]西北农林科技大学信息工程学院,陕西杨凌712100

出　　处：《农业机械学报》2015年第3期152-157,共6页Transactions of the Chinese Society for Agricultural Machinery

基　　金：'十二五'国家科技支撑计划资助项目(2012BAH29B00)

摘　　要：为实现对土壤速效磷含量的快速测定,以关中塿土为材料,研究基于光谱分析的土壤速效磷含量测定方法。首先用便携式近红外频谱仪在不同采样高度下,采集土壤样本900-1 700 nm波长范围的漫反射光谱,采用3倍标准差准则和主成分分析得分图对异常样本进行判别和剔除,然后对比分析4种波长选择方法对建模效果的影响,发现基于稳定的竞争性自适应加权抽样法的结果最佳,最后通过分析不同非线性建模方法对预测结果影响实验,探明最小二乘支持向量机方法的预测结果最好。实验结果表明,采样高度为10 cm时本文建立模型的土壤速效磷含量预测决定系数为0.858 1,均方根误差为10.880 1,具有较高的精度,可对土壤速效磷含量进行快速预测。The aim of this research is to realize the rapid measurement of soil available P content. The suitable proportion of available P could promote the crops grow. Taking ＇Lou＇ soil as sample, the soil diffusion reflectance spectrum in 900 ~ 1 700 nm under different observation heights were collected by usingthe portable spectrographs. Firstly, five observation heights （5, 7, 10, 12, 15 cm） were compared, and 10 cm was considered to be the best. The abnormal samples were identified and eliminated by using 3 times standard deviation and principal component analysis method. That effectively improved the model precision. Then, the effect of four different wavelengths selecting methods （SPA, CARS, sCARS, RF） on modeling was analyzed. The result showed that sCARS was the best. Finally, the different nonlinear modeling methods （RBF neural network, WNN, LSSVM） were experimented. The results proved that LSSVM had the best result. When the observation height was 10 era, the modeling prediction correlation coefficient was O. 858 1, and the prediction root mean square error was 10. 880 1. The results showed a high accuracy and feasibility of soil available P content prediction.

关 键 词：土壤 速效磷 光谱分析 预测模型 SCARS 最小二乘支持向量机 

分 类 号：S153.61[农业科学—土壤学] O657.33[农业科学—农业基础科学]