表面损伤和可溶性固形物是丰水梨品质的重要指标, 为降低丰水梨的在线分选伤果率, 探索丰水梨可溶性固形物套网分选的可行性。 在运动速度5个·s-1、 积分时间80 ms的条件下, 采集350~1 150 nm范围内的丰水梨套网前后的近红外漫透射光谱。 对比分析了丰水梨套网前后的光谱特征差异, 套网检测的光谱有效信息降低。 为提高套网检测光谱的有效信息, 利用多项式拟合的方法剥离网套光谱。 并采用丰水梨套网前后、 多项式拟合剥离网套光谱后的光谱建立可溶性固形物在线检测模型。 对所建立的模型进行对比分析, 其中采用二次多项式拟合剥离后的光谱建模效果最佳, 建模集均方根误差为0.328°Brix, 相关系数为0.95, 最终实现了丰水梨可溶性固形物的套网在线检测。 采用未参与建模的36个样品进行在线分选准确性评价, 分选准确性为94.4%。 试验结果表明, 研究可为含水量高、 易碰伤、 易擦伤的水果提供在线分选策略及理论依据。

Subsurface damage and soluble solid content (SSC) are important indexes for evaluating the quality of hosui pear. The feasibility was discussed for nondestructive detecting with net for SSC of intact hosui pear by online visible-near infrared (visible-NIR) transmittance spectroscopy. Ten tungsten halogen lamps were installed in a sorting line. The total power of lamp was 1 000 watt. The spectra were recorded with the integration time of 80 ms in the wavelength range of 600~900 nm when the samples were conveyed at the speed of five samples per second. The light sources were illuminated from both sides of the production line, and the detector received light from the bottom of the fruit cup. The spectrum of each sample was recorded automatically by use of the hardware trigger mode. The index plate and driving gear were mounted on the same shaft. The location of the index plate's tooth was matched with the location of the fruit cup. Hall sensor was placed at a height of 2 mm above the tooth of the index plate. When the index plate turned one tooth, a Hall sensor sent a 3.5 V high frequency signal to trigger spectrometer to save one spectrum. The response properties of visible-NIR spectra for normal pears and pears with net were analyzed. The effective information of the spectra of pears with net was reduced, and the new method of polynomial fitting was used to eliminate this problem. The influence of pears with net on SSC determination was also explored by partial least square (PLS) regression models, and PLS model was employed with normal pears and pears with net. The robust model was developed by the spectra which eliminates the influence of net by binomial fitting. Therefore, a new strategy was proposed for detection of SSC of intact pears by online visible-NIR transmittance spectroscopy. The new samples, which were not used in the calibration, were used to access the abilities of predicting SSC of intact pears. The accuracy of sorting grade was 94.4% according to the SSC values and the SEP was 0.328°Brix of predicting SSC. The results showed that nondestructive detecting with net for SSC of intact hosui pear was feasible by visible-NIR transmittance spectroscopy. The research can provide technical support and reference basis for grading of mass fruits quality.