样品温度对近红外光谱有很大影响, 在近红外技术评价水果品质的实际应用时, 需要修正温度变化对模型预测结果的影响。便携式近红外光谱仪采集不同温度下（0～30 ℃）苹果的漫透射光谱, 采用二阶导数和卷积平滑进行预处理。选取20 ℃下代表性样本的光谱数据, 建立基准PLS模型。斜率/偏差法分别计算苹果糖度PLS模型在0, 10和30 ℃下的修正方程。分析结果表明: 斜率/偏差法对0, 10和30 ℃下外部样本预测结果进行修正, 预测精度得到显著提高, 其修正前后的Q值分别为0.525 cv 0.810, 　0.680 cv 0.822, 0.669 cv 0.802。温度修正模型可以有效提高预测精度, 也扩展了近红外仪器的适用性, 为自主研发便携式近红外光谱仪提供参考。

Visible　(Vis)/near infrared (NIR) spectroscopy has been used successfully to measure soluble solids content (SSC) in fruit. However, for practical implementation, the NIR technique needs to be able to compensate for fruit temperature fluctuations, as it was observed that the sample temperature affects the NIR spectrum. A portable Vis / NIR spectrometer was used to collect diffused transmittance spectra of apples at different temperatures (0～30　℃). The spectral data of apple at 20　℃ was used to develop a norm partial least squares (PLS) model. Slope/ bias technique was found to well suits to control the accuracy of the calibration model for SSC concerning temperature fluctuations. The correctional PLS models were used to predict the SSC of apple at 0, 10 and 30　℃, respectively. The correctional method was found to perform well with Q values of 0.810, 0.822　and 0.802, respectively. When no precautions are taken, the Q value on the SSC may be as small as 0.525～0.680. The results obtained highlight the potential of portable Vis/NIR instruments for assessing internal quality of fruits on site under varying weather conditions.