摘  要：针对室外光照对近红外光谱检测带来误差的问题，提出基于模型传递来减少检测误差的方法。以圆黄梨为样品，分析样品在室内、室外阴影下的近红外光谱，建立室内光谱的偏最小二乘（PLS）模型。采用直接校正（direct standardization，DS）算法，减小室内外光谱差距，使得室内PLS模型能预测室外光谱。结果表明：在室内建立的模型能预测经DS算法传递后的室外光谱，预测决定系数（r2p）和预测均方根误差（root mean square error of prediction，RMSEP）分别为0.71和0.374，能有效解决室外光照对光谱检测的影响。

关键词：近红外光谱；直接校正；光照影响；模型传递；糖度

文献标志码：A       文章编号：1674-5124（2015）12-0070-04

Research on model transfer of near infrared spectroscopy at different illumination

ZHANG Wenjun， TANG Hong， JIANG Qiaoyong

（College of Metrological Technology and Engineering，China Jiliang University，Hangzhou 310018，China）

Abstract: To minimize the error in sample detection by near infrared spectrometers outdoor， a method using model transfer to lower the measurement error has been proposed in this paper. The near infrared spectroscopy （NIRS） was used to detect the sample-Wonhuwang pear under outdoor and indoor shadows and the results were analyzed to establish a Partial Least Squares （PLS） model for indoor spectrum. The detection errors between the near infrared spectrums under indoor and outdoor shadows were reduced through Direct Standardization （DS） algorithm to enable the model to predict the near infrared spectrum outdoor. Experimental results show the determination coefficient and root mean square error of prediction are 0.71 and 0.374 respectively. It can lower the effect of outdoor light on spectrum detection.

Keywords: NIRS； DS； illumination； model transfer； sugar