Integration of Hyperspectral Imaging and Chemometrics for Internal Quality Evaluation of Packaged and Non-Packaged Fresh Fruits

Research on packaged fruits has seen a notable upturn primarily driven by consumers’ desire for fruit safety and quality across the distribution network. This study examined the effectiveness of hyperspectral imaging (HSI) combined with chemometrics to assess the internal quality of packaged and non-packaged fresh fruits. Visible–near-infrared (Vis-NIR; 400–1000 nm) and short-wave infrared (SWIR; 1000–2500 nm) hyperspectral images of apples and plums were captured using 200 samples for each fruit across three groups—plastic wrap (PW), polyethylene terephthalate (PET) box, and non-packaged (NP)—for the prediction of soluble solid content (SSC), moisture content (MC), and pH. A partial least square regression (PLSR) model demonstrated promising results on SSC and MC across all sample groups in both Vis-NIR and SWIR, with performance ranked NP > PW > PET. Calibration and prediction coefficients of determination (R2) exceeded 0.82, 0.80, and 0.79, with root mean square errors (RMSE) less than 0.57, 0.59, and 0.59 for NP, PW, and PET, respectively. This research outcome confirmed the suitability of HSI as a critical instrument for predicting the composition of fresh fruits inside plastic packaging, offering a quick and non-invasive approach for quality evaluation in supply chains.