Application of shortwave near infrared spectroscopy in determining physicochemical properties of jackfruit

Jackfruit (Artocarpus heterophyllus) is a tropical fruit which belongs to Moraceae family. Jackfruit has a very good market in Malaysia for fresh consumption. The fruit have also been exported to Asian and European countries due to its premium quality. In order to ensure that only jackfruits at the best quality are exported, the quality of the fruits should be monitored at an orchard regularly. The development of an efficient and rapid measurement technique to assess the quality (sweetness, moisture content and nutritive values) of jackfruit non-destructively are critically needed in the industry. Therefore, this research aimed to investigate the potential application of shortwave near infrared spectroscopy (SWNIRS) to predict the physicochemical properties of jackfruit. In this study, three typical physicochemical properties of jackfruit such as Soluble Solids Content (SSC), pH and moisture content (MC) were measured. A total of 29 fresh jackfruits samples were used in this investigation, giving a total of 870 skin and 798 pulp portions. The SSC and MC of the skin and pulp portions were recorded using a handheld digital refractometer and conventional hot air-drying, respectively. While the pH values of the pulp were measured using ionic pH meter. The spectral data of the both portions were recorded using SWNIRS with the wavelength ranged from 500 to 950 nm. Partial least square (PLS) regression analysis was chosen to establish regression models between the spectral data and the quality parameters. Useful information from the spectral data were extracted by Principal component analysis (PCA). The pre-processing methods, PLS and PCA exercises were run using Unscrambler X 10.3 software to evaluate the performance of the models.From the results, it was found that the values of R² and root means square errors for calibration (RMSEC) in predicting SSC from the skin samples were 0.77 and 0.90 °Brix, respectively. For the prediction model of the same portions, the values of R² and root mean square errors of prediction (RMSEP) were 0.69 and 0.97 °Brix, respectively. For the pulp portions, it was found that the values of R² and RMSEC were 0.92 and 1.79 °Brix, respectively. In terms of prediction model, the values of R² and RMSEP were 0.76 and 3.19 °Brix, respectively. The ability of the spectrometer in predicting MC from the both skin and pulp portions were also investigated. The results showed that R² and RMSEC values in predicting MC from the skin samples were 0.65 and 2.18%, respectively. For the prediction model of the same portions, the values of R² and RMSEP were 0.69 and 2.81 %, respectively. For the pulp portions, it was found that the values of R² and RMSEC were 0.83 and 2.27%, respectively. In terms of prediction model, the values of R² and RMSEP were 0.68 and 3.07%, respectively. In addition to SSC and MC, the potential use of the spectrometer in predicting pH values from the pulp samples was also investigated. From the results, it was found that the values of R² and RMSEC in predicting pH values form the pulp samples were 0.92 and 0.16, respectively. For the prediction model of the same portions, the values of R² and RMSEP were 0.82 and 0.23, respectively. Overall, it is concluded that the SWNIRS has the potential to be used for predicting the physicochemical properties of jackfruit from the skin and pulp portions. The development of rapid and portable SWNIRS will be very helpful for farmers in the jackfruit industry.

Text FK 2019 155 - ir.pdf Download (1MB)

Actions (login required)

View Item