Cutlery Detection System With Camera and Yolov5

The culinary business industry, such as restaurants, heavily relies on consumer experience. One aspect that can affect consumer experience is the service provided by the restaurant in supplying complete dining equipment. Misunderstandings may occur due to the lack of complete dining and drinking utensils in restaurants, leading to complaints from customers or guests. However, manually checking the completeness of dining equipment can be time-consuming and can easily lead to human error. Therefore, it is necessary to develop a system that capable to detect the completeness of dining equipment on restaurant tables. To address this, this research uses cameras and machine learning to build a system that can automatically recognize dining equipment on the dining table. The machine learning model used in this research is YOLOv5. In YOLOv5, two processes occur such as training and testing. The training process is carried out on an annotated dataset to produce a trained model. Then this model is tested in the testing process to evaluate its performance using a separate dataset different from the training dataset. The model run on Google Colab showed an accuracy rate of 90% during training and 93.75% during testing.