Deteksi dan Klasifikasi KendaraanMenggunakan OpenCV pada Rekaman Video ATCS di Kota Palembang

Abstract

The Area Traffic Control System (ATCS) is designed to monitor and regulate traffic flow in real time by processing data captured from CCTV cameras at intersections. This study presents the development of an automated vehicle detection and classification system using OpenCV (Open Source Computer Vision Library) on ATCS video recordings. The system aims to count and classify vehicles to generate quantitative traffic data and support peak-hour analysis. The results show that the proposed computer vision–based system provides a practical solution to assist traffic monitoring and decision-making in urban transportation planning. Based on the testing results at several intersections in Palembang City, it was found that the most dominant vehicle categories are cars and motorcycles, with the highest percentages occurring during peak hours in the morning and evening. The distribution of vehicles indicates that central areas such as Walikota Intersection and Palembang Indah Mall experience the highest traffic volumes, while peripheral areas such as Tanjung Api-Api are relatively less congested. This condition reflects the high mobility of Palembang residents on weekdays, particularly in commercial and government areas. Overall, the implementation of this automated detection system provides quantitative insights into traffic density and flow patterns, which are useful for optimizing data-driven traffic management.