Rapid and Non-Destructive Detection of Decay in Peach Fruit at the Cold Environment Using a Self-Developed Handheld Electronic-Nose System

Abstract

Non-destructive detection of decay in fruit in real time during cold chain is important to recycle the decayed fruit in time for reverse supply chain. Fruit is commonly stored inside external packages during cold chain, making neither manual observation nor optical inspection techniques available to detect the decay in fruit. In this work, the potential of a self-developed handheld electronic nose (e-nose) instrument to non-destructively acquire volatile substances and then detect decay in peach fruit during cold chain (0 °C) was explored. A desktop e-nose instrument was considered as a comparison. The storage days of peach fruit during storage were also predicted by two instruments. Partial least squares discriminant analysis and least squares support vector machines (LS-SVM) were used for the classification of decay in peach fruit. Partial least squares regression and LS-SVM were used for the prediction of the storage days. Successive projection algorithm (SPA), uninformation variable elimination (UVE), UVE-SPA, and competitive adaptive reweighted sampling were applied to select the characteristic variables from e-nose data. The best model for the classification of decayed fruit during cold chain by the handheld e-nose instrument had the correct answer rate of prediction of 95.83% (94.64% for healthy samples and 100.00% for decayed samples). The best model for predicting the storage days of peach fruit during cold chain by the handheld e-nose instrument had the residual predictive deviation value of 9.283. The results indicate that the self-developed handheld e-nose system is a simple and non-destructive tool to detect decay in peach fruit during cold storage.