基于机器视觉与图卷积网络的矿区无人驾驶车辆异常行为检测

Abnormal Behavior Detection of Unmanned Vehicle in Mining Area Based on Machine Vision and Graph Convolutional Network

无人驾驶车辆在矿山逐步得到关注与应用的同时,也存在一定的安全隐患。为提高车辆异常行为检测效率,提出了一种基于机器视觉与图卷积网络的矿区无人驾驶车辆异常行为检测方法。该方法通过分析矿区无人驾驶车辆行驶轨迹,利用机器视觉技术对车辆进行识别、跟踪和检测。首先获取无人驾驶车辆图像监测数据,进而进行特征提取,包括颜色、形状、大小等关键特征,并据此对车辆进行识别、跟踪和检测。其次,使用图卷积网络对车辆行为进行建模,将车辆驶过的路段视为图中的节点,不同路段之间的连接视为边,通过图卷积网络学习车辆的行为,并对车辆异常行为进行检测。试验结果表明:该方法可以在短时间内准确检测到矿区无人驾驶车辆的异常行为,并对异常情况及时报警,有助于提高矿山生产的安全性和效率。

While unmanned vehicles have gradually been paid attention to and applied in mines,there are also certain safety risks.In order to improve the monitoring efficiency of vehicle abnormal behavior,a detection method of abnormal behavior of unmanned vehicle in mining areas based on machine vision and graph convolutional network was proposed.The method analyzes the trajectory of unmanned vehicle in mining areas,and uses machine vision technology to identify,track and monitor the vehicles.Firstly,the image monitoring data of the unmanned vehicle is obtained,and then the feature extraction is carried out,including the key features such as color,shape and size,and the vehicle is identified,tracked and detected accordingly.Secondly,the graph convolutional network is adopted to model the vehicle behavior,and the road section passed by the vehicle is regarded as the node in the graph,and the connection between different roads is regarded as the edge.The graph convolutional network is adopted to learn the vehicle behavior and monitor the abnormal behavior of the vehicle.The experimental results show that the method can accurately detect the abnormal behavior of the unmanned vehicle in the mining area in a short time,and timely alarm the abnormal situation,which is helpful to improve the safety and efficiency of the mine production.