The rapid growth of urbanization has made traffic management and road safety critical, with congestion, infrastructure deterioration, and human errors leading to accidents, injuries, and fatalities. Factors like fatigue, distraction, and drowsiness further increase risks, emphasizing the need for intelligent transportation systems (ITS) to enhance safety and efficiency. Deep learning has revolutionized object detection, enabling real-time models to recognize complex patterns in images and videos. Advanced models like Real-Time Detection Transformer (RT-DETR) and YOLO improve detection speed and accuracy, allowing autonomous systems to swiftly process visual data, identify hazards, and make informed driving decisions.

Асфау С.К. (науч. рук. Шаветов С.В.) Research on algorithms of computer vision for safety motion control of self-driving cars // Сборник тезисов докладов конгресса молодых ученых. Электронное издание. – СПб: Университет ИТМО, [2025]. URL: https://kmu.itmo.ru/digests/article/15623