基于深度神经网络的驾驶员违章行为识别方法

Driver Illegal Behavior Recognition Method based on Deep Neural Network

驾驶员的违章行为是造成交通事故的主要原因之一,利用摄像头实时监控行驶过程中的驾驶员违章行为是一个减少交通事故的有效方法.本文提出一种通过深度神经网络的驾驶员违章行为识别方法.首先,利用Deep-Pose检测驾驶员身体关键点,接着,基于这些关键点提取动作特征,然后使用最近邻分类器识别典型违章行为.实验证明,本文的方法对于典型的违章行为是有效的.

Driver's illegal behavior is one of the main causes of traffic accidents.Using camera to monitor driver's illegal behavior in real time is an effective method to reduce the number of traffic accidents.This paper proposed a driver's behavior recognition meth od based on deep neural network.First,the driver’s body key points are detected by Deep-Pose method.Then,the action features are extracted based on these key points.Finally,the nearest neighbor classifier method is used to identify typical illegal behavior.Experiments show that the method is effective for typical Driver's violations.