Designing of a Smart Workspace Video Surveillance System Isyeri Uygulamalari i in Akilli bir Video G zetim Sistemi Tasarimi

AKBAŞ A., Alghabawi A.

33rd IEEE Conference on Signal Processing and Communications Applications, SIU 2025, İstanbul, Turkey, 25 - 28 June 2025, (Full Text) identifier

  • Publication Type: Conference Paper / Full Text
  • Doi Number: 10.1109/siu66497.2025.11111855
  • City: İstanbul
  • Country: Turkey
  • Keywords: computer vision, smart video surveillance systems, video processing, workplace surveillance
  • Bilecik Şeyh Edebali University Affiliated: Yes

Abstract

A smart workplace video surveillance system has been designed to monitor employees' working performance. The system classifies the per-second activity status of employees by using body positions generated by MediaPipe Pose as input data. The Random Forest (RF) model used for this purpose was trained and tested by a generated dataset with 17482 elements. The results of 5-fold cross-validation tests show that designed system provides a generalizable performance over 98%.