Vehicle type detection based on deep learning in traffic scene

Abstract Nowadays, vehicle type detection plays an important role in the traffic scene. Deep Learning algorithm has been widely used in the field of object detection. Applied the Faster RCNN framework, improved the RPN networks, which was an effective and representative of the Convolutional Neural Network of deep learning on object classification algorithm, and combined with the MIT and Caltech car dataset as well as some different types of vehicle pictures in the Internet, to detection and recognization the three types of vehicles which are commom in traffic scene. The experimental results show the effectiveness and high-efficiency of method of deep learning in the vehicle type detection.