People Counting in Computer Rooms Using YOLO Algorithm at Office of Academic Resources and Information Technology, Rajamangala University of Technology Suvarnabhumi

Abstract

The purposes of this research were: 1) to develop a system for counting the number of entries and exits of users of a computer room; 2) to evaluate the performance of the developed system; and 3) to investigate user satisfaction with the system. The system was developed in Python on the Windows operating system and employed the YOLOv8 algorithm to detect and count people from webcam video streams. Two sample groups were used in the study: 1) a system performance testing group, consisting of students appearing in real entry–exit events during actual computer room service hours, selected through incidental sampling (49 students for camera position 1 and 64 students for camera position 2); and 2) a user satisfaction group, consisting of 23 personnel from the Educational Technology and Information Technology divisions, selected through purposive sampling. The research instruments were the developed entry–exit counting system and a user satisfaction questionnaire. Data were analyzed using descriptive statistics, namely mean and standard deviation.

The results showed that 1) the entry–exit counting system for the computer room of the Office of Academic Resources and Information Technology, Rajamangala University of Technology Suvarnabhumi, was successfully developed and comprised seven functional components; 2) the system performance testing indicated that the camera installation position affected detection accuracy, with the side-view camera position yielding the highest accuracy 100% accuracy for detecting entries and 98.44% accuracy for detecting exits; and 3) overall user satisfaction with the developed system was at the highest level (  = 4.55, S.D. = 0.56).