【摘要】 交通标志的自动识别对于导航电子地图数据采集和更新效率的提高有着重要的应用价值,通过对交通标志影像的特征分析,在RGB模型的基础上,使用自适应分量域值进行图像的归一化,能快速准确地抽取交通标志特征影像,采用多向颜色直方图最小二乘法特征匹配进行特征识别,可以有效地识别出变形、缩放、部分遮挡及污脏的交通标志,通过C~#编程进行相关图像处理算法的实现,在对实际的交通标志影像的自动识别试验中取得了良好的效果。更多还原

【Abstract】 Road database consists of an inventory of features,topology and attributes.But these information have changed fre- quently with the development of roads and traffic regulations.This is a severe problem to the digital maps’ providers.The old methods for gathering geo-information can not satisfy the markets demands,so there is a requirement to build a new and intelli- gent system to get accurate and reliable map databases,and to integrate most of the current surveying techniques,i.e.GPS,in- ertial sensors,digital imagery,laser ranging.This article mainly discusses how to use the digital image technology to auto-rec- ognize traffic signs in MDCS (Mobile Data Capturing System).With multilevel image process to extract their feature and match them,the veracity of recognizing traffic signs can be improved.Therefore,increasing efficiency of traffic sign surveying and up- dating the period of surveying can be achieved.更多还原