Abstract: When students leave class at the same time and go to the cafeteria for lunch, the cafeteria is crowded and students need to wait in a long queue. In response to this situation, a smart restaurant system construction plan is proposed, which uses artificial intelligence and Internet of Things technology on a system integration platform. Using binocular recognition technology, infrared cameras collect information about people entering and leaving the restaurant, and transmit it to a cloud server to calculate the number of people entering and leaving the restaurant, the total number of diners, and the number of empty seats. The information is transmitted to the terminal display screen for display. Using the KMS take-out cabinet system, students order meals online, and the cafeteria prepares meals in advance according to demand. Students take out their meals offline. If students order meals on-site at the restaurant, select dishes, and place them in a visual settlement system, the system will automatically identify and quote the dishes. Diners can pay by face recognition or QR code payment. The construction and application of smart restaurants not only solve the problem of crowded dining queues for students at the same time, but also save manpower, material resources, and reduce inventory for school cafeterias, achieving the goal of improving restaurant service quality.