基于Python的车辆检测算法的实现  

Realization of Vehicle Detection Algorithm Based on Python

作　　者：张巍[1] Zhang Wei(Taiyuan Branch of China Joint Network Communication Co.,Ltd.,Taiyuan Shanxi 030001)

机构地区：[1]中国联合网络通信有限公司太原市分公司

出　　处：《机械管理开发》2019年第12期258-261,共4页Mechanical Management and Development

摘　　要：近年来,我国汽车保有量不断增加,交通事故成为一个需要高度重视的话题。因此,辅助驾驶系统的重要性不断提高,而车辆防碰撞系统是辅助驾驶系统的重要组成,其中前车检测是防碰撞系统的关键模块。鉴于此,使用Python编程语言实现了基于机器学习的车辆检测算法,算法的主要步骤如下,首先将彩色图片转换为灰度图像,直方图均衡化,中值滤波,然后提取出该区域的Haar-like特征并使用预先训练好的分类器进行分类,从而实现车辆检测的功能。本文完成了预定的目标,可以成功标记出一幅图片中的车辆。In recent years,the number of cars in China has been increasing,and traffic accidents have become a topic that needs to be attached great importance to.Therefore,the importance of auxiliary driving system is increasing,and vehicle anti-collision system is an important component of auxiliary driving system,in which front vehicle detection is the key module of anti-collision system.In view of this,the vehicle detection algorithm based on machine learning is implemented by using Python programming language.the main steps of the algorithm are as follows:firstly,the color picture is converted into gray image,the histogram is equalized and the median filter is filtered,and then the Haar-like features of the region are extracted and divided by a pre-trained classifier.And thus the function of vehicle detection is realized.The purpose of this paper is to successfully mark the vehicle in a picture.

关 键 词：车辆检测 机器学习 HAAR-LIKE特征 ADABOOST分类器 

分 类 号：U270.67[机械工程—车辆工程]