YOLOv11 for Vehicle Detection: Advancements, Performance, and Applications in Intelligent Transportation Systems

Vehicle detection is a critical component in the development of advanced intelligent transportation systems (ITS), which rely on accurate and real-time information to optimize traffic flow, enhance safety, and support autonomous vehicle technologies [1]. As the number of vehicles on the road continues to grow, the demand for robust vehicle detection systems capable of operating under varying conditions--such as changes in weather, lighting, and vehicle types--has become paramount. In traffic monitoring, vehicle detection enables the real-time analysis of traffic patterns, congestion management, and incident detection, contributing to more efficient urban mobility. Moreover, vehicle detection serves as the foundation for vehicle classification and tracking systems, which are essential for dynamic tolling, traffic law enforcement, and infrastructure planning [2]. The evolution of vehicle detection systems has been closely tied to advancements in deep learning, particularly in the field of convolutional neural networks (CNNs) [3]. CNNs have played a pivotal role in object detection tasks due to their ability to automatically learn hierarchical features from raw image data [4, 5, 6, 7, 8]. Traditional vehicle detection approaches, such as histogram of oriented gradients (HOG)[9] and support vector machines (SVM)[10], lacked the flexibility and scalability needed for modern applications, especially when dealing with complex scenes and varying environmental conditions. Subsequent methods, like Scale-Invariant Feature Transform (SIFT) and Speeded-Up Robust Features (SURF) [11], introduced improvements in detecting and describing features under varying scale and rotation conditions, although computational constraints limited real-time applicability for ITS.