Using Machine Learning Algorithm to Determine Food Expiration Status in Smart Refrigerators

This study investigates the application of Convolutional Neural Networks (CNNs) for real-time food quality assessment and inventory management in smart refrigeration systems. Three pre-trained CNN architecturesVGG19, Xception, and ResNet50-were evaluated using image data from the Flickr API to classify food items as fresh or spoiled. VGG19 demonstrated superior performance with a validation accuracy of approximately 98%, while Xception exhibited signs of overfitting. Furthermore, a custom CNN model was developed to assess stock depletion using userprovided images, revealing that the incorporation of Dense layers significantly enhanced classification accuracy from 77% to 87%. These findings underscore the potential of CNN-driven automation in optimizing food preservation and supply monitoring within intelligent refrigeration systems.