【摘要】 [目的]利用近红外光谱技术建立预测南疆核桃水分的检测模型。[方法]通过对南疆温185核桃进行近红外光谱水分无损检测,并用标准烘干法对所建立的检测模型进行验证,形成测定核桃水分的快速检测模型。[结果]试验表明,对采集的光谱数据进行SNV预处理,并采用偏最小二乘回归分析法(PLS)建立模型,得到了较小的SEC值和较高的RC值。其SNV预测的值用标准烘干法得出的平均偏差为0.35%,完全满足对南疆温185核桃进行水分无损检测的要求,可以应用到实际的核桃水分预测当中。[结论]研究可为核桃水分的快速无损检测提供参考依据,同时将此项无损检测技术应用到果农的生产实际当中,可提高果农的收益。更多还原

【Abstract】 [Objective]To establish detection model for forecasting water in walnut in south Xinjiang by NIRS. [Method]Research is based on the nondestructive testing of 185 types of walnuts from southern Xinjiang using NIRS( Near Infrared Spectroscopy) in a moist environment.The calibration model was validated using a standard drying method. [Result]The results demonstrate that the spectral data collected for SNV pretreatment,using partial least squares regression analysis( PLS) as the model,resulted in smaller SEC values and higher RC values. The SNV average deviation of the predicted value,obtained using standard drying method,was 0. 35%. This fully satisfies the requirements of nondestructive moisture inspection for all 185 Southern Xinjiang walnut varieties,and it can be used as a method to predict the actual level of moisture in walnuts. [Conclusion]The study can provide reference basis for rapid and nondestructive testing of walnut water. The technology applied in actual production can improve fruit growers benefit.更多还原