Rapid determination of cabbage quality using visible and near-infrared spectroscopy

- The first use of Vis/NIR spectroscopy to determine the quality of fresh cabbage.
- Rapid and non-destructive measurement for cabbage quality inspection.
- Different mode of measurements affects the prediction accuracy of determination.
- Vis/NIR (500-1100 nm) was optimal for partial linear regression square models.

The feasibility of Vis-NIR spectroscopy was investigated for its ability to determine the quality of cabbage. This study aims to compare the prediction ability of interactance and reflectance measurements in determining the quality of cabbage (moisture, SSC and ascorbic acid contents). The wavelength ranges of 500-1100 nm were used in this experiment. Two types of data preprocessing were applied to enhance the calibration model based on partial least squares (PLS) regression with respect to the logarithms of reciprocal absorbance (log (1/R)), its first and second derivatives. The PLS regression models for moisture content yielded correlation coefficients (R2) of 0.48-0.67 and root mean square error of prediction (RMSEP) of 2.34-2.83 g/kg for interactance, with R2 of 0.58-0.74 and RMSEP of 2.50-3.25 g/kg for reflectance. The PLS statistics for SSC were R2 of 0.59-0.66 and RMSEP of 0.20-0.22 °Brix for interactance and R2 of 0.53-0.64 and RMSEP of 0.22-0.27 °Brix for reflectance, respectively. Statistics for ascorbic acid content were R2 of 0.24-0.61 and RMSEP of 0.11-0.15 g/kg FW for interactance and R2 of 0.35-0.38 and RMSEP of 0.13 g/kg FW for reflectance. Finally, it is possible to use the Vis/NIR spectroscopy as a rapid tool for evaluating the cabbage quality.