基于YOLOv5+Deep-SORT的运煤车辆目标检测与跟踪

Target Detection and Tracking of Coal Vehicles Based on YOLOv5+Deep-SORT

煤的运输对山西煤矿资源的管理有着重要意义。本研究使用某洗煤厂的入口监控视频进行间隔帧的提取,选取mAP@0.5为0.957的YOLOv5算法对视频中运煤车辆进行目标检测,在此基础上使用Deep-SORT算法进行目标跟踪,并实现了运煤车辆的统计。系统的设计和实现解决了洗煤厂在运煤过程中煤丢失的问题。

The transportation of coal is of great significance to the management of coal resources in Shanxi.In this study,the interval frames are extracted from the video of entrance monitoring of a coal washery,and the relevant data sets are made.mAP@0.5 of YOLOv5 algorithm is used to detect the target of coal vehicles in the video.On this basis,the Deep-SORT algorithm is used to track the target,and the statistics of coal vehicles are realized.The design and implementation of the system solves the problems of coal loss in the process of transporting coal.