Search

In this study, a portable optical instrument was designed and developed for real-time nondestructive determination of fruit ripeness. A prototype of the instrument using inexpensive parts was developed and calibrated for apple ripeness detection based on moisture content, soluble solids content, pH, and firmness. An array of six optical point sources (light-emitting diodes (LEDs) that emit light at 520-, 670-, 920-, 970-, 1050-, and 1320-nm wavelengths) were selected using visible/near infrarred (Vis/NIR) spectral analysis on apple samples. The instrument was equipped with optical fibers, optical sensors, and an Arduino board for sending and receiving the waves and processing the received signals, as well as controlling the operation of the instrument. The decision tree method was utilized to create a classification model for training the instrument to classify ripe and unripe apples. Test results showed acceptable performance of the developed instrument for apple ripeness estimation and proved useful in classifying fruits into ripe and unripe groups based on their internal quality. [ABSTRACT FROM AUTHOR]