Self-checkout Intelligent Recognition System for Fruit Supermarkets

In response to issues such as product barcode contamination and prolonged checkout waiting times during customers’ shopping process at supermarkets, this paper proposes a method for automatic identification of fruits and vegetables based on image recognition technology. The system utilizes TensorFlow to train two models, CNN and MobileNet, and compares their accuracy on the validation set. Ultimately, the MobileNetV2 model, which achieves higher accuracy, is selected to accomplish automatic fruit recognition. This checkout system can enhance customers’ shopping experience, improve sales efficiency, and reduce labor costs.