论文部分内容阅读
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80ms的条件下,采集350~1 150nm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。
Surface damage and soluble solids are important indexes of pear quality. In order to reduce the rate of on-line sorting and injury of pear, we explored the feasibility of the selection of soluble solid nets. Near-infrared diffuse transmittance spectra were collected before and after the Fengmurei pear nets in the range of 350-1 150 nm under the conditions of 5 s -1 and 80 ms integration velocity. The differences of spectral characteristics before and after Fengmian pear were compared and analyzed, and the spectral effective information of nets detection was reduced. In order to improve the effective information of the detection spectrum of the nets, the spectra of the nets were stripped by means of polynomial fitting. Before and after using Pearlescent Pearson, the model of on-line detection of soluble solids was established by the polynomial fitting spectrum after peeled off the nets. The comparative analysis of the established model, in which the use of quadratic polynomial fitting after stripping the spectral modeling the best effect, the modeling set root mean square error of 0.328 ° Brix, the correlation coefficient of 0.95, and ultimately fruiting pear Soluble solid Net sets of online testing. 36 samples not involved in modeling were used to evaluate the accuracy of on-line sorting, the sorting accuracy was 94.4%. The test results show that the study can provide online sorting strategies and theoretical basis for the fruit with high water content, easy bruising and easy bruising.