Analysis of Motor Vehicle Density Based on Digital Images (Case Study on Jenderal Sudirman Street, Palembang City)

Abstract

Digital image processing has become an essential technology in motor vehicle density analysis, particularly in real-time traffic monitoring. The ability of this system to produce accurate visual data enables researchers and transportation planners to identify traffic patterns, predict congestion, and make data-driven decisions. This study has two main objectives: to determine how motor vehicle density on Jalan Jenderal Sudirman, Palembang City, is analyzed based on digital images and to identify the most appropriate digital imaging method to support this analysis. The Haar Cascade method was chosen as the main algorithm due to its reliability in feature-based object detection. The process begins with image pre-processing to reduce noise and improve lighting in order to enhance image quality. Next, image segmentation is performed to separate vehicle objects from the road background. Detection is performed frame-by-frame, enabling the system to detect vehicles quickly and respond to traffic dynamics. The detected data is then processed using a simple calculation algorithm to determine the number of vehicles within a certain time frame, which is then used to analyze traffic density levels. The results of the study show that the Haar Cascade method is capable of detecting and counting vehicles with an accuracy of 83.6%, making it an efficient solution in intelligent traffic monitoring systems. This research makes a real contribution to the field of modern transportation, particularly in the development of digital image-based systems to support more measurable and technology-based urban transportation planning and decision-making.