近红外漫反射光谱法在稻谷质量检测中的应用

摘要

284 samples of paddy rice in Jiangxi,China,were collected by near-infrared spectroscopy DA7200.And the chemi-calcompositionanalysiswasconductedinlinewiththerequirementofrelevantnationalstandardmethods.Basedonthechemi-calanalysis,ananalyticalmodelwasestablishedbypartialleastsquares(PLS)method.Andtheaccuracyofthemodelwas verified afterwards.As a result,higher correlation was found in the quantitative analysis models developed by PLS.Another 30 selected samples were tested by the model.The result indicated that both the correlation coefficient and root mean square error prediction was able to meet the relevant national standards'requirements.In this regard,the near-infrared diffuse reflectance spectroscopy is available to the rapid non-destructive testing for quality assessment of the paddy rice.%采用DA7200型二极管阵列近红外光谱仪对江西地区的284份稻谷样品进行光谱采集，同时按照国家标准方法对试验样品进行化学分析检验。在化学分析检验的基础上，采用偏最小二乘法建立分析模型，并对其准确性进行了评价探讨。结果显示，采用偏最小二乘法所建立的定量分析模型相关性较高。同时，另选取30份样品进行模型验证，其相关系数和预测均方根误差均符合国家标准方法要求，因此近红外漫反射光谱法可以用于稻谷质量评定中的快速无损检测。