应用圆形度和颜色直方图的交通信号灯识别  

Traffic light recognition with circularity and color histogram

作　　者：谷明琴[1] 蔡自兴[1] 李仪[1] 

机构地区：[1]中南大学信息科学与工程学院,湖南长沙410083

出　　处：《计算机工程与设计》2012年第1期243-247,共5页Computer Engineering and Design

基　　金：国家自然科学基金项目(90820302;60805027);国家博士点基金项目(200805330005);湖南省院士基金项目(20010FJ4030);广东省自然科学基金项目(94512001002983)

摘　　要：针对智能车辆介绍了一种在城市环境中识别交通信号灯的新方法。该方法先对原图像进行预处理,过滤不符合形态学准则的区域。计算候选区域的圆形度和背板的颜色信息,把圆形度符合和背板为黑色的区域作为交通信号灯区域。并将检测到的交通信号灯区域图像从RGB空间转换到HSV空间中,用颜色直方图对图像的H分量进行分布统计,根据其分布特征来识别交通信号灯的类型。实验结果表明,该方法能有效地识别复杂环境下的交通信号灯。For intelligent vehicle, a new method is introduced to recognize traffic lights in urban environments. Firstly, an original image is preprocessed, removed the regions which don＇ t fit to key criteria of shapes of image. Then, candidate regions＇ circularities and background color are computed and detected; traffic light region is demarcated by the detected result. Finally, the traffic light region images which are detected converted color space form RGB to HSV, the number of hue （H） of labeled region images are counted by color histogram, as basis to judge the traffic light type, The experiments show that our algorithm is useful to recognize traffic light under complex environment.

关 键 词：交通信号灯识别 圆形度 颜色直方图 HSV颜色空间 智能车辆 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]