|
Abstract
|
Vehicle information detection plays a vital role in the development of intelligent transportation systems (ITS) [1]. These systems enhance transportation efficiency and safety by utilizing sensing, analytical, control, and communication technologies within ground transportation networks. ITS applications collect and transmit data to alleviate traffic congestion, optimize traffic management, and respond more swiftly to unforeseen events. In modern surveillance and security frameworks, identifying and tracking vehicle movement is essential across multiple sectors, including law enforcement and smart city initiatives. Extracting key vehicle attributes and features contributes to enhanced public safety and effective traffic management [2,3].
Vehicle characteristics extraction has gained significant attention due to its broad application in areas such as criminal investigations, toll collection, license plate recognition, vehicle counting, incident detection, and traffic monitoring. Accurately identifying vehicle types, license plates, logos, and colors enhances law enforcement efforts, traffic management, and parking control. In vehicle-related crimes, police often rely on detailed descriptions provided by victims to search through extensive traffic image databases for a suspect vehicle. Recognition of license plates, along with inherent features like vehicle type, logo, and color, plays a crucial role in improving vehicle identification and search accuracy [4].
In recent years, digital image processing algorithms have been widely adopted for estimating vehicle attributes such as shape, color, speed, and position. This progress has been further supported by the availability of more powerful computing resources. High-core processors and graphics processing units (GPUs) have become standard components in computer vision-based systems, enabling real-time object recognition and classification. The integration of deep learning and machine learning techniques into computer vi
|