Search

Spectral data of the outer wall and inner capsule of Pericarpium Citri Reticulatae Viride from different regions (Anhui, Guangdong, and Sichuan) were collected with portable near infrared spectrometer. Multiple interferences in the spectra were eliminated using single and combined pretreatment methods. Pattern recognition methods such as principal component analysis (PCA), cluster independent soft pattern classification (SIMCA) and Fisher linear discriminant analysis (FLDA) were used to establish traceability models of Pericarpium Citri Reticulatae Viride origin. The results showed that spectral pretreatment could eliminate the interferences of baseline drift, background, and spectral peak overlap to a certain extent. However, the traceability analysis of the origin couldn't be achieved. Among the three pattern recognition methods, accurate traceability analysis of Pericarpium Citri Reticulatae Virid couldn't be achieved with PCA method. The whole identification rate of SIMCA model with the outer wall and inner capsule original spectra were 99.14% and 98.28%, respectively. The whole identification rates of FLDA models were both 99.57%, better than that of SIMCA models. The identification rates of SIMCA and FLDA models with appropriate spectral pretreatment were 100%. Therefore, portable near infrared spectroscopy technology combined with the supervised pattern recognition methods can achieve non-destructive traceability analysis of Pericarpium Citri Reticulatae Viride origin, expanding a new way for the traceability analysis of food and drug homologous substances origin. [ABSTRACT FROM AUTHOR]

The nondestructive characterization of citrus varieties (Egyptian sweet orange, Lane Late navel orange, Australian orange, and Blood orange) was developed based on near-infrared diffuse reflectance spectroscopy (NIRDRS) together with principal component analysis (PCA) and Fisher linear discriminant analysis (FLDA). An experiment for the penetration of NIRDRS into the peel was designed and the effects of different spectral acquisition points were investigated. Pretreatments were used to eliminate the spectral interferences. As an unsupervised pattern recognition method, PCA was used to establish the characterization models. Furthermore, supervised pattern recognition based on PCA and FLDA was employed to enhance the accuracy. The results of the penetration experiments show that near-infrared light enters the citrus peel and is able to characterize the internal composition. Even with the optimized spectral pretreatment, accurate characterization of citrus varieties was not achieved by PCA. However, the accurate characterization of citrus varieties was provided by PCA-FLDA. The accuracies of four spectral acquisition points are 95%, while the characterization accuracies of six spectral acquisition points are 100% combined with optimized spectral pretreatment. Therefore, NIRDRS with PCA-FLDA is suitable for the rapid and nondestructive characterization of citrus varieties. [ABSTRACT FROM AUTHOR]

An accurate and nondestructive detection method of hidden mold infection in citrus was established based on portable near infrared diffuse reflectance spectroscopy and chemometric methods. [Display omitted] • Nondestructive detection of hidden mold infection in citrus was achieved. • NIRDRS light was demonstrated to penetrate the citrus peel. • Models were validated by the validation and external validation sets. • Identification models of LWNIR were much better than those of SWNIR. • Identification accuracy of LWNIR-DT-PLS-DA model was 100%. An accurate and nondestructive detection method of hidden mold infection in citrus was established based on portable near infrared diffuse reflectance spectroscopy (NIRDRS) and chemometric methods. Penetrability of NIRDRS light on the peel of Chunjian hybrid citrus was studied. The results show that NIRDRS light can penetrate the peel to a certain extent, while the penetrability of short-wave near infrared (SWNIR) was better than that of long-wave near infrared (LWNIR). The identification models of hidden mold infection were established with five pattern recognition methods combined with different wavelength bands. The results show that the identification models of LWNIR were much better than those of SWNIR. 100% identification accuracies of LWNIR were obtained with soft independent pattern classification (SIMCA), support vector machine (SVM), partial least squares discriminant analysis (PLS-DA) and the optimized pretreatment methods. In addition, the developed models were further validated by the external validation set collected one month later. [ABSTRACT FROM AUTHOR]