面向驾驶辅助系统的车辆与行人安全检测方法研究

Research on security detection methods of vehicle and pedestrian for driving assistance system

智能化交通系统解决方案是处理交通问题的有力手段。目前,已经有越来越多的行人检测方法被提出,多种检测方法已被应用到汽车驾驶辅助系统中。针对目前车辆与行人的安全检测方法存图像的特征大多依靠人工提取得到,需要不断尝试来得到最适合的特征,成本较高的问题,采用对比的方法对传统行人及车辆检测技术进行分析,提出基于深度学习和卷积神经网络的方式构造出分类器来处理车辆的检测问题的应用,得出基于深度学习方式和卷积神经网络的方式构造出分类器,更方便构建出车辆检测系统。

Intelligent traffic system solution is a powerful means to deal with traffic problems.Nowadays,more and more pedestrian detection methods have been proposed,many detection methods have been applied to the vehicle driving assistance system.In the current safety detection methods of vehicles and pedestrians,most of the image features are extracted by manual extraction,and continuous attempts are required to obtain the most suitable features with high cost.It analyzes the traditional pedestrian and vehicle detection techniques by comparison,and proposes to deal with vehicle detection problems based on deep learning and convolutional neural network(CNN),which is more convenient to construct the vehicle detection system.