Real-Time Table Occupancy Detection in Restaurants Using Object Detection and Computer Vision Techniques

Authors : Ali Kerem Güler, Ali Musa

Pages : 12-27

View : 124 | Download : 94

Publication Date : 2025-06-27

Article Type : Research Paper

Abstract :This work presents a new approach to monitoring and analyzing table occupancy in a restaurant setting using object detection algorithms. The method involves creating a custom image dataset of tables of different colors, shapes, and sizes, and training a model on this dataset using the YOLO (You Look Only Once) algorithm. The system is designed to detect tables and calculate occupancy measurements based on the number of people detected in the relevant area around each table. In addition, information including table occupancy is recorded via logging in a time series dataset format to facilitate future operational planning and time-based analysis. In the preliminary tests, the number of individuals seated at the table was manually determined by reviewing camera recordings for a specific time interval. Subsequently, a comparison was made between this manual count and the automated detection performed by the system. The results of this comparison revealed that the system accurately detected the number of people seated at the table during the specified time interval. By saving and analyzing this data, enterprises can make informed operational decisions and improve their service quality to increase customer satisfaction.
Keywords : Nesne Tanıma, Görüntü İşleme, Bilgisayarlı Görü, Yolov8

ORIGINAL ARTICLE URL