In-depth understanding of the shifting of water status during dehydration is crucial for obtaining better quality of dried food. In this work, we report a nondestructive method to measure the water status in hot-air dried mushroom via multispectral imaging (MSI) technology combined with chemometric methods. The low-field nuclear magnetic resonance (LF-NMR) measurements were performed as reference. During drying process, the moisture content changed dramatically with notable migration and conversion of different water phases. Partial least squares (PLS), back propagation neural network (BPNN), and least squares-support vector machine (LS-SVM) models were applied to develop quantitative models. Among all, BPNN model showed considerably better performance of prediction with coefficient of determination R2 c = 0.9829, R2 p = 0.9639. The results demonstrated that MSI technology combined with chemometric methods is an impressive approach for determination of the water status in hot-air dried mushrooms, which would facilitate infield of food processing by providing applicable and appropriate platform. PRACTICAL APPLICATION: Experimental investigation of different water status during food processing. Assessment of the potential of multispectral imaging to predict water status. Usage of novel measurement method for food processors.

中文翻译：

---

用于预测热空气脱水过程中蘑菇水分状态的多光谱成像

深入了解脱水过程中水分状态的变化对于获得更好的干燥食品质量至关重要。在这项工作中，我们报告了一种通过多光谱成像 (MSI) 技术结合化学计量学方法测量热风干蘑菇中水分状态的无损方法。进行低场核磁共振 (LF-NMR) 测量作为参考。在干燥过程中，水分含量发生显着变化，不同水相发生显着迁移和转化。应用偏最小二乘 (PLS)、反向传播神经网络 (BPNN) 和最小二乘支持向量机 (LS-SVM) 模型来开发定量模型。其中，BPNN 模型显示出相当好的预测性能，决定系数 R2 c = 0.9829，R2 p = 0.9639。结果表明，MSI 技术与化学计量学方法相结合是确定热风干蘑菇中水分状态的一种令人印象深刻的方法，它将通过提供适用和适当的平台来促进食品加工领域的发展。实际应用：食品加工过程中不同水分状态的实验研究。评估多光谱成像预测水状况的潜力。用于食品加工商的新型测量方法的使用。评估多光谱成像预测水状况的潜力。用于食品加工商的新型测量方法的使用。评估多光谱成像预测水状况的潜力。用于食品加工商的新型测量方法的使用。