结合先验知识和图像特征的道路提取方法  

Combined prior knowledge and image characteristics for lane detection method

作　　者：林丽群[1] 肖俊[2] 

机构地区：[1]湖北大学资源与环境学院,武汉430062 [2]武汉大学测绘遥感信息工程国家重点实验室,武汉430079

出　　处：《计算机工程与应用》2011年第33期236-239,243,共5页Computer Engineering and Applications

摘　　要：致力于在复杂环境下能对多种道路进行检测,提出了一种先验知识库与自适应区域增长相融合的道路检测方法,通过少量的道路样本采集,建立样本库,训练挖掘出道路知识模型,并结合区域增长方法对分割的实时道路影像进行道路区域增长。实验结果表明,该方法适用于多种不同环境道路的提取,鲁棒性强。In the complex environment,this paper presents a lane detection method with the ability to deal with both structured and unstructured roads.The method combines a priori knowledge database with adaptive region growing method.It establishes a priori knowledge database through a set of road sampling,training,data mining and knowledge discovery,combines adaptive road region growing method using this database process on real-time segmented images.The experimental results show that this method is able to robustly find the road area on different types of roads in various environments.

关 键 词：道路检测 计算机视觉 区域增长 

分 类 号：TP751[自动化与计算机技术—检测技术与自动化装置]