红外与毫米波学报
紅外與毫米波學報
홍외여호미파학보
JOURNAL OF INFRARED AND MILLIMETER WAVES
2010年
1期
32-37
,共6页
沈掌泉%王珂%Xuewen Huang
瀋掌泉%王珂%Xuewen Huang
침장천%왕가%Xuewen Huang
近红外光谱%土壤碳含量%行走式测定%波段算术组合%偏最小二乘回归法
近紅外光譜%土壤碳含量%行走式測定%波段算術組閤%偏最小二乘迴歸法
근홍외광보%토양탄함량%행주식측정%파단산술조합%편최소이승회귀법
near-infrared spectroscopy%soil carbon content%on-the-go measurement%band arithmetic combination%partial least squares regression(RLSR)
以田间行走式设备获取的近红外光谱数据为基础,利用最小二乘回归法(PLSR)建立了应用近红外光谱数据预测土壤碳含量的校正模型,与利用原始光谱数据建立的模型相比,应用经比值或归一化差值处理的光谱数据建立的校正模型可以提高预测精度.精度提高的原因可能是光谱数据经过波段算术组合处理后,能降低模型建立过程中产生过配的风险,使模型能包括更多的成分和信息.研究结果表明,利用偏最小二乘回归法,可以有效地建立田间近红外光谱与土壤碳含量之间的校正模型;同时,应用比值或归一化差值这些波段算术组合方法来处理近红外光谱数据,可以进一步提高模型的预测精度.因此,应用行走式设备获取的近红外光谱数据来快速测定田间土壤中碳的含量是可行的.
以田間行走式設備穫取的近紅外光譜數據為基礎,利用最小二乘迴歸法(PLSR)建立瞭應用近紅外光譜數據預測土壤碳含量的校正模型,與利用原始光譜數據建立的模型相比,應用經比值或歸一化差值處理的光譜數據建立的校正模型可以提高預測精度.精度提高的原因可能是光譜數據經過波段算術組閤處理後,能降低模型建立過程中產生過配的風險,使模型能包括更多的成分和信息.研究結果錶明,利用偏最小二乘迴歸法,可以有效地建立田間近紅外光譜與土壤碳含量之間的校正模型;同時,應用比值或歸一化差值這些波段算術組閤方法來處理近紅外光譜數據,可以進一步提高模型的預測精度.因此,應用行走式設備穫取的近紅外光譜數據來快速測定田間土壤中碳的含量是可行的.
이전간행주식설비획취적근홍외광보수거위기출,이용최소이승회귀법(PLSR)건립료응용근홍외광보수거예측토양탄함량적교정모형,여이용원시광보수거건립적모형상비,응용경비치혹귀일화차치처리적광보수거건립적교정모형가이제고예측정도.정도제고적원인가능시광보수거경과파단산술조합처리후,능강저모형건립과정중산생과배적풍험,사모형능포괄경다적성분화신식.연구결과표명,이용편최소이승회귀법,가이유효지건립전간근홍외광보여토양탄함량지간적교정모형;동시,응용비치혹귀일화차치저사파단산술조합방법래처리근홍외광보수거,가이진일보제고모형적예측정도.인차,응용행주식설비획취적근홍외광보수거래쾌속측정전간토양중탄적함량시가행적.
Partial least squares regression (PLSR) was employed to build predicting model of the content of soil carbon with on-the-go near-infrared reflectance spectroscopy (NIRS) measurements. The model based on band ratio or normalized difference of NIRS data can improve the prediction precision than the model with the original NIRS data. The reasons might be that the process of band arithmetic combination could reduce the risk of overfitting and it made the model include more useful components and information. The results show that the effective calibration model between field NIRS and the content of soil carbon can be set up by PLSR, and predicting precision can be improved while band arithmetic combination of ratio or normalized difference is performed on the NIRS data before modeling. Thus, it is feasible to estimate the content of soil carbon quickly in the field by on-the-go NIRS measurement.