Robustness of models based on NIR spectra for sugar content prediction in apples

摘要: The sugar content of Golden Delicious apples is predicted using near infrared (NIR) spectrometry. study focuses on the metrological characteristics measurement and external parameters involved in lack robustness NIR-based model. were fruit temperature, spectrometer temperature ambient light. first two factors influenced prediction accuracy: (i) a variation altered prediction, relationship seems to be described by non-linear model within considered range, (ii) also linear function for between 4 30°C. Ambient light did not show have any influence analysis showed satisfactory repeatibility with low error, 0.073°Brix. reproducibility was good regarding bias-corrected standard error (SEPc) without significant differences experiments, other hand bias remained even if previous maintained constant. These results will taken into account future measurements, order improve developed apples.