基于LED组合光源的水晶梨可溶性固形物和大小在线检测
Online detection of soluble solids content and size of crystal pear based on LEDs light source-detector
-
摘要: 该文探讨了近红外光谱技术结合发光二极管（LED）组合光源探头在线检测水晶梨的可溶性固形物和大小的可行性。试验中采用850、880和940 nm 3盏LED组成组合光源探头，每个水晶梨在均匀成单列的输送线上以每秒5个梨的速度运动，采用漫反射方式采集水晶梨的漫反射光谱。应用偏最小二乘（PLS）和最小二乘支持向量机（LS-SVM）方法建立了可溶性固形物和大小2个理化参数的校正模型，同时对不同光谱预处理方法（平滑、一阶微分、二阶微分）建立的模型的预测性能进行了对比分析，并通过外部验证来检验模型预测的准确性。利用平滑处理光谱建立偏最小二乘（PLS）模型的预测效果最优，可溶性固形物和大小的相关系数分别为0.86和0.90，预测均方根误差分别为0.58%和 1.93 mm。试验研究表明：应用近红外光谱技术结合LED组合光源探头在线检测水晶梨可溶性固形物和大小具有可行性。Abstract: Online detection method of soluble solids content (SSC) and size of crystal pear using LEDs light source-detector based on near infrared spectroscopy was studied. LEDs light source-detector with wavelengths of 850 nm, 880 nm and 940 nm were used to irradiate crystal pear in this experiment. The crystal pear was homogeneously arranged on conveyor line at the speed of 5 pears per second. Spectra was measured in near infrared diffuse reflectance mode. Three pre-processing methods including average smoothing, first and second derivatives were applied to improve the predictive ability of the models. Partial least squares (PLS) and least squares support vector machine (LS-SVM) were used to develop calibration models. The prediction set was used to evaluate the predictive ability of the models. The results showed that the best model was obtained by PLS with the pre-processing method of average smoothing. The correlation coefficient (R) and root mean square error of prediction (RMSEP) was (0.86, 0.58%) and (0.90, 1.93 mm) for soluble solids content and size, respectively. The results showed that online detection of soluble solids content and size of crystal pear based on near infrared diffuse reflectance spectroscopy combined with LEDs light source-detector was feasible.
-
计量
- 文章访问数: 1665
- HTML全文浏览量: 0
- PDF下载量: 953
下载:
