摘要
针对复杂交通环境下的ADB汽车大灯检测挑战,提出了基于改进YOLOv5算法的解决方案。通过对YOLOv5算法进行优化,融合特征融合、核心网络及视野拓展层等先进技术,实现了对车辆行驶环境的精准检测。实验结果表明,改进后的YOLOv5算法在速度、每秒帧数(FPS)和参数量上均表现出显著优势,检测精度得到大幅提升。同时,结合扩展卡尔曼滤波技术,有效预测了目标车灯光源的轨迹,进一步增强了系统的鲁棒性和实用性。不仅为ADB汽车大灯的环境检测提供了新的思路和方法,也为智能车灯控制系统的未来发展奠定了坚实基础,有助于提升道路行驶的安全性和智能化水平。
A solution based on the improved YOLOv5 algorithm is proposed to address the challenges of ADB car headlight detection in complex traffic environments.By optimizing the YOLOv5 algorithm and integrating advanced technologies such as feature fusion,core network,and field of view expansion layer,precise detection of vehicle driving environment has been achieved.The experimental results show that the improved YOLOv5 algorithm exhibits significant advantages in speed,frames per second(FPS),and parameter count,resulting in a significant improvement in detection accuracy.At the same time,combined with extended Kalman filtering technology,the trajectory of the target car light source is effectively predicted,further enhancing the robustness and practicality of the system.Not only does it provide new ideas and methods for environmental detection of ADB car headlights,but it also lays a solid foundation for the future development of intelligent headlight control systems,which helps to improve the safety and intelligence level of road driving.
作者
郑雅伟
Zheng Yawei(Shanxi Institute of Economics and Business,Taiyuan Shanxi 030024,China)
出处
《山西电子技术》
2025年第2期123-126,共4页
Shanxi Electronic Technology
作者简介
郑雅伟(1984-),男,山西交口人,讲师,硕士,研究方向:电子与通信工程。
