Design of intelligent fire control system based on machine vision

In the intelligent control of cooking robots, heat regulation is the core aspect that affects the quality of dishes. In traditional cooking processes, fire control relies on manual experience, which often leads to issues such as uneven doneness and low efficiency. This paper designs an intelligent automatic fire control system based on machine vision. The system captures images of ingredients through a camera and utilizes the YOLOv5 deep learning model to identify the doneness of ingredients, including four states: raw, medium, fully cooked, and over cooked. Combined with embedded control technology, it drives a servo motor to adjust the power of an induction cooker, forming a complete closed-loop control process. In the experimental section, stir-fried chili and pork is selected as an example, and an automatic stirring robot is introduced for variable control. The system’s automatic heat adjustment is compared with manual heat adjustment. The experiment shows that the system achieves an average accuracy rate of 95% in doneness recognition, enabling precise and efficient control of heat, significantly superior to manual control, with a response time of less than 1 second. The system achieves full automation of the cooking process, providing a practical solution for the development of smart kitchen equipment for households.