Intelligent Transportation System Research Based on ITS-YOLO

With the rapid development of Intelligent Transportation Systems (ITS), accurate detection of traffic signs, pedestrians, and vehicles is crucial for enhancing road safety and transportation efficiency. This paper proposes an intelligent transportation system based on an improved YOLOv11 model, optimized for the complexity of traffic scenes. Specifically, an Illumination-Guided Attention Block (IGAB) is introduced to enhance detection accuracy for small targets. Experimental results demonstrate that the proposed method achieves a 4.8% improvement in mAP50 and a 15.2% increase in inference speed compared to the original YOLOv11 on the China Traffic Sign Detection Benchmark (CCTSDB 2021), also achieving significant improvements on a self-collected dataset. ITS-YOLO ensures real-time performance while maintaining high accuracy. Compared to traditional YOLO architectures, the proposed method significantly enhances performance in detecting small-scale traffic signs, reducing both false positives and missed detections. This system can be widely applied in autonomous driving, intelligent surveillance, and other fields, providing more efficient and precise visual perception capabilities for intelligent transportation.