The non-destructive method for detection of fungal contamination in peach fruit using hyperspectral imaging was evaluated. Growth characteristics of three major spoilage fungi in peach fruit during decay were estimated. Three quantitative prediction models were then constructed to forecast the microbial content from the HSI datasets. The prediction of fungal contamination on the fruit was visualized with different colors. Additionally, principal component analysis (PCA) was applied to reduce the dimensionality of the HSI data and to discriminate the infection degree in peaches. The results showed that partial least squares regression (PLSR) could achieve performance with R_p² not less than 0.84in predicting fungal colony counts, while PCA scores successfully identified the infected degrees of samples. This study illustrates that HSI combined with chemometrics can potentially be implemented for the quantitative detection of fungal contamination in peach fruit.

评估了使用高光谱成像技术检测桃果实中真菌污染的非破坏性方法。估算了桃果实腐烂过程中三种主要腐败真菌的生长特性。然后构建了三个定量预测模型，以从HSI数据集中预测微生物含量。水果上真菌污染的预测以不同的颜色显示。此外，应用主成分分析（PCA）来减少HSI数据的维数并区分桃子中的感染程度。结果表明，偏最小二乘回归（PLSR）可以达到R _p ^2的性能。预测真菌菌落数不低于0.84，而PCA评分成功地确定了样品的感染程度。这项研究表明，将HSI与化学计量学结合可以潜在地用于定量检测桃果实中的真菌污染。