Manufacture and performance test of banana ripe detection tool using laser light backscattering imaging

Postharvest technology for the detection of banana quality and ripeness is developing rapidly, especially non-destructive technology (NDT). However, some NDT technologies require long detection times and expensive instruments. One of the optical-based methods, namely laser light backscattering imaging (LLBI), has recently become popular because it is fast, reliable, effective, economical, and real-time. The purpose of this study was to design and test the LLBI image acquisition system, which has good accuracy in predicting the ripeness of Cavendish bananas. A system has been designed with dimensions of 80 cm×80 cm×80 cm, an angled iron frame equipped with a bridge jack, a 650 nm 5 mW laser, a camera with a focal length of 18–105 mm, and a laptop equipped with digiCamControl software and Fiji ImageJ. The system built has been tested through statistical calculations and obtained >62% of LLBI parameters have a strong correlation with the maturity level. Correlation can build a model with a suitability of 99.44% in predicting the maturity level. From this model, we also obtained 99.42% of the LLBI parameters that affect the maturity level. In addition, direct performance measurement was carried out with the results of 87.63% of the modeling results by real conditions. All test results illustrate that the LLBI system can predict the ripeness of Cavendish bananas with good accuracy.