Non destructive testing method for lamb shelf life based on multimodal data fusion

<b><sc>Abstract.</sc></b> Flexible sensing technology has become an emerging assessment tool in the field of food quality testing in recent years, and has promising applications in the detection of lamb products. The aim of this study is to investigate the potential of detecting the effective shelf life of lamb meat through multimodal data fusion techniques. In this study, the multimodal data of lamb meat at different storage temperatures were collected by flexible sensors, and the modal data of lamb meat were analyzed for correlation, and the information fusion technique was used to establish a model for lamb meat quality detection. Comparing the model AlexNet network model, GoolgeNet network model, and ResNet-50 network model established in this study, the model established in this study achieves an accuracy of 94.12% and 92.33% in the training set and the test set, respectively, which can better detect the quality of lamb meat. The results of this study show that the information fusion technology provides a strong technical support for the effective shelf-life detection of lamb products, which is expected to promote the expansion of the application of this technology in the field of food quality detection.