A primary study on forecasting the days before decay of peach fruit using near-infrared spectroscopy and electronic nose techniques

Forecasting the number of days until peach fruit decay is important not only for consumers to determine when to eat the fruit, but also for sellers to determine their sale strategies. However, traditional visual observation, chemical and anatomy-digital caliper methods are applicable only when the decay has already begun. In this work, the possibility of forecasting the days before decay (DBD) of peach fruit was explored by means of near-infrared (NIR) spectroscopy and an electronic nose (e-nose). Partial least squares regression, least-squares support vector machines, and multiple Gaussian fitting regressions were used for model calibration. Successive projections algorithm, uninformation variable elimination, and competitive adaptive reweighted sampling were used for variable selection. The best DBD prediction model had a correct answer rate of 82.26%. The results show that the combination of NIR spectroscopy and e-nose data holds promise as a reliable and rapid alternative to forecasting the DBD of peach fruit. This study reveals the attractive prospect of non-destructively estimating how long peach fruit can be edible before decaying, which is important for improving both the daily lives of people and management efficiency in the peach industry.