LED-induced fluorescence spectroscopy technique for apple freshness and quality detection

• A novel and compact LED-induced fluorescence system was constructed.
• 375 nm LEDs are more effective than the 400 nm LEDs for apple quality detection.
• Accuracy of freshness and quality predictions is better than obtained with NIR.

The LED-induced fluorescence spectroscopy technique is developed to detect apple freshness and quality. A compact and automatic system is set up for fast measurements of apple fluorescence spectra. The partial least square regression (PLSR) method is used to establish a predictive model between fluorescence signals and corresponding actual apple qualities. Apple freshness is related to physiological aging, and apple quality is expressed by firmness (F) and soluble solids content (SSC). As for freshness detection, an LED at 375 nm is demonstrated to be more effective than at 400 nm with a root mean square error of validation (RMSEV) of 4.73 d. Meanwhile, the prediction results of the firmness and SSC for two kinds of apples also show low RMSEV values of 6.88 N and 0.98%, respectively, indicating that the LED-induced fluorescence spectroscopy technique is an effective, convenient and promising method for scanning apples and perhaps other foods.