基于数字图像的汽车驾驶员行驶状态判别  

Drivers Working States Recognition Based on Digital Images

作　　者：林广宇[1] 魏朗[1] 

机构地区：[1]长安大学汽车学院,西安710064

出　　处：《计算机工程》2007年第22期193-194,197,共3页Computer Engineering

基　　金：博士学科点专项科研基金资助项目(20050710002)

摘　　要：通过车载CCD图像传感器摄取图像,在利用中值滤波、边缘检测等图像处理技术去除噪声和获取道路标线的基础上,建立了摄像机的透视投影模型和汽车驾驶员行驶状态模型,研究了车辆行驶过程中相对道路标线的行驶状态参数,以监控驾驶员行车状况。实验证明,该方法获得的行驶状态参数曲线能有效判别驾驶员的行驶状态,为减少驾驶员人为因素导致的交通事故作了有益的探索。This paper discusses the vehicle traveling parameters related to the road traffic marking to monitor drivers working states at real time, With the help of images taken by an on-board CCD image sensor. Based on suppressing noise and detecting the position of road traffic marking in the images by using image processing techniques such as median filter and edge detection, it establishes models for perspective projection and drivers working states recognition. Experimental results show that the approach is effective in judging driving condition and it is important to cut down traffic accidents caused by drivers＇ personal factors.

关 键 词：图像分析 道路标线 驾驶员行驶状态 交通事故 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]