|
|||||
|
|
| 基于小波变换的近红外光谱预测土壤有机质 | |
| 引用本文： | 杨苗,左月明,杨萍果.基于小波变换的近红外光谱预测土壤有机质[J].山西农业大学学报(自然科学版),2010,30(2):154-158. |
| 作者姓名： | 杨苗 左月明 杨萍果 |
| 作者单位： | 1. 山西农业大学,文理学院,山西,太谷,030801 2. 山西农业大学工程技术学院,山西,太谷,030801 3. 山西师范大学,生命科学院,山西,临汾,041000 |
| 基金项目： | 山西农业大学科技创新基金，山西省基础研究项目 |
| 摘 要： | 为了探索快速检测土壤有机质含量的方法,试验采用不同分解水平的Coiflet函数的小波(wavelet)分析方法,对山西关帝山土壤样品的近红外光谱信号进行了消噪处理,来快速获取土壤中有机质含量。结果表明:对有机质敏感波段的为450~600 nm,810~935 nm,1 030~1 315 nm,1 380~1 400 nm;有机质NIRS法与实验室标准法测定值之间的相关系数R2为0.9818;说明通过小波变换滤波,选择敏感波段,用偏最小二乘回归方法预测土壤有机质含量是可行的。 |
| 关 键 词： | 小波变换 土壤光谱 有机质 偏最小二乘回归 |
Prediction of Soil Organic Matter with Near-infrared Spectroscopy Treated by Wavelet Transformation |
|
| YANG Miao,ZUO Yue-ming,YANG Ping-guo.Prediction of Soil Organic Matter with Near-infrared Spectroscopy Treated by Wavelet Transformation[J].Journal of Shanxi Agricultural University,2010,30(2):154-158. | |
| Authors: | YANG Miao ZUO Yue-ming YANG Ping-guo |
| Abstract: | In order to explore the rapid detection method of soil organic matter content,this paper adopts a different decomposition levels of wavelet Coiflet function(wavelet) analysis method.Soil samples from Shanxi Guandi Mountain were carried out near-infrared spectroscopy signal de-noising processing,to quickly obtain the soil organic matter content.The results showed that sensitive bands to the organic matter were 450 nm~600 nm,810 nm~935 nm,1 030 nm~1 315 nm and 1 380 nm~1 400 nm;The correlation coefficient between the value of organic matter NIRS method and the standard laboratory method was 0.9818;Through wavelet filtering,sensitive band selection employing partial least-squares regression method to predict soil organic matter content is feasible. |
| Keywords: | Wavelet transformation soil spectra organic matter partial least squares regression |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ | |