Robust non-destructive prediction of color attributes associated with bronzing in jackfruit (Artocarpus heterophyllus cv. ‘Tekam Yellow’) using visible–near infrared spectroscopy

This study investigated the potential of visible–near-infrared spectroscopy (Vis–NIRS) as a non-destructive tool for predicting color attributes associated with internal bronzing in jackfruit (Artocarpus heterophyllus cv. ‘Tekam Yellow’), a disorder associated with Pantoea stewartii subsp. stewartii. Bronzing affects the pulp without visible external symptoms, leading to postharvest losses. Diffuse reflectance spectral data (500–950 nm) were collected from intact fruit at 10, 12, and 14 weeks after anthesis (WAA), and partial least squares regression (PLSR) models were developed to predict rind and flesh color parameters (L*, a*, b*, C*, ∆E, and h°). Spectral preprocessing methods, including Savitzky–Golay smoothing, standard normal variate (SNV), and multiplicative scatter correction (MSC), were applied to improve model performance. The optimized models showed strong calibration and prediction accuracy, with coefficients of determination (Rc2 and Rp2) up to 0.98 and low prediction errors (root mean square error of prediction as low as 0.29). High predictive performance was observed under normal conditions, particularly for rind and flesh L* at 10 WAA (Rp2 ≥ 0.95; RPD > 10). In contrast, models for bronzed fruit showed reduced predictive accuracy, indicating increased spectral variability in diseased tissues. Vis–NIRS demonstrated strong potential as a rapid and non-invasive technique for assessing internal quality and supporting indirect assessment of bronzing-associated physiological changes in jackfruit. This approach offers a promising tool for improving postharvest quality control and supporting sustainable management in the jackfruit industry.

Actions (login required)

View Item