Vehicle types classification using deep neural network techniques

Abstract

Abstract-Traf ic low is one of the most important information elements in intelligent traf ic transportation engineering. This study developed a vehicle type classi ication system using a neural network technique. The architecture of this study is divided into two parts, vehicle pictures are collected irst, and then divided into motorcycles, sedans, recreation vehicles, buses and trucks to build a contrast database. The image processing techniques included median iltering and edge detection used to de-noise to improve recognition ef iciency. The second stage is processing the previous data stage into the system identi ication database. All data created by the database were then input into the classi ier for calculation.