| 基于OSC和PLS的土壤有机质近红外光谱测定 |
| |
| 引用本文： | 宋海燕,何勇. 基于OSC和PLS的土壤有机质近红外光谱测定[J]. 农业机械学报, 2007, 38(12): 113-115,189 |
| |
| 作者姓名： | 宋海燕 何勇 |
| |
| 作者单位： | 1. 山西农业大学工程技术学院 030801 太谷县
2. 浙江大学生物系统工程与食品科学学院 310029 杭州市 |
| |
| 基金项目： | 国家自然科学基金;浙江省科技厅资助项目;山西农业大学博士科研基金 |
| |
| 摘 要： | 分析了经过简单处理的土壤样本光谱特性,将正交信号校正与偏最小二乘算法回归相结合,建立了土壤光谱特性与土壤有机质含量之间的定量分析模型。结果表明,正交信号校正可以消除噪声信息对土壤有机质含量预测的影响,预测样本的预测相关系数达到0.893、标准偏差为0.051%、预测标准差为0.050%;而不采用正交信号校正建立定量分析模型的对应参数分别为0.818、0.069%和0.085%。
|
| 关 键 词： | 土壤有机质 近红外光谱 偏最小二乘算法 正交信号校正 |
| 收稿时间： | 2007-02-05 |
| 修稿时间： | 2007-02-05 |
Near Infrared Determination of Organic Matter Content in Soil Based on OSC and PLS |
| |
| Song Haiyan,He Yong. Near Infrared Determination of Organic Matter Content in Soil Based on OSC and PLS[J]. Transactions of the Chinese Society for Agricultural Machinery, 2007, 38(12): 113-115,189 |
| |
| Authors: | Song Haiyan He Yong |
| |
| Affiliation: | 1.Shanxi Agricultural University 2.Zhejiang University |
| |
| Abstract: | Spectral properties of simply treated soil samples were analyzed by using Nicolet intelligent Fourier transform (FT) infrared spectrum. A new pretreatment method-orthogonal signal correction (OSC) was presented to eliminate the influence of the noise on soil organic matter (SOM) content prediction. Partial least square (PLS) analysis has been used to build prediction models with calibration data of 67 samples. The remaining 20 samples were used to validate the models. The result showed that OSC-PLS could improve the prediction ability greatly. The correlation coefficient is 0.893, standard error of prediction (SEP) is 0.051%, and root mean standard error of prediction (RMSEP) is 0.050% respectively. |
| |
| Keywords: | SOM Near infrared spectroscopy PLS OSC |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
