改进的目标检测算法在司机室场景的应用

Application of Improved Object Detection Algorithm in Driver’s Cab Scene

针对司机室智能分析系统中的智能检测模块进行优化改进,提升检测算法在复杂场景下的准确率。提出一种YOLOv8-DR方法,在YOLOv8的基础上引入坐标注意力机制,增加一个大尺度检测头分支,更好地捕捉玩手机等小目标的细节信息。通过实验对比,该方法提升13.5%的准确率,可实时预警规范司机驾驶行为,保证机车安全运行。

In this paper,the intelligent detection module in the intelligent analysis system of the driver’s cab is optimized and improved to improve the accuracy of the detection algorithm in complex csenarios.This paper proposes a YOLOv8-DR method,which introduces coordinate attention mechanism on the basis of YOLOv8,and adds a large-scale detection head branch to better capture the details of small targets such as playing mobile phones.Through experimental comparison,the accuracy of this method is improved by 13.5%,and the driver’s driving behavior can be warned in real time to ensure the safe operation of the locomotive.