Research on Optimization strategy of YOLOv8 target detection model based on scene adaptation

This study focuses on the target detection task in the automatic driving scene. Aiming at the deficiency of the traditional YOLOv8 model in complex scene adaptability, a scenario-based YOLOv8 target detection model is proposed. Through the design of scene classification module, multi-branch YOLOv8 model and dynamic parameter adjustment strategy, combined with data enhancement, model fusion and optimization technology, the detection accuracy and generalization ability of the model in different scenarios are effectively improved. In the experiment, BDD100K data set was used for training and testing, and the results showed that the method significantly improved the detection performance in bad weather and low light environment, and provided strong support for the development of autonomous driving technology.