Improved CNN Automatic Identification Algorithm for Food Foreign Bodies Based on LeNet-5 Network

The purpose of this study was to improve the speed and accuracy of foreign body identification in food. An improved Convolutional Neural Network (CNN) model was constructed by adding a batch normalization layer and a dropout layer to the LeNet-5 network architecture, and a recognition system was built to realize the automatic identification of foreign objects in food images. Experimental results show that compared with the traditional model, the improved model has significant advantages in detection accuracy and recognition speed, and its recognition accuracy for food foreign bodies can reach 99.75%, and the recognition time is only 0.332 seconds, indicating that the constructed dumpling image foreign body recognition model has excellent performance in detection speed and recognition accuracy, and has a good application prospect.