基于线段和消隐点的建筑墙面自动分割算法

建筑物的墙面分割是基于图像建模、路标识别、导航及场景理解等室外城市环境应用系统中的基础内容,目前基于多幅图像重建的墙面分割方法或基于重复结构假设的先验知识的分割方法具有计算量大和通用性差等特点。为了能够简捷、快速、准确地得到墙面分割结果,提出一种针对建筑物的墙面自动分割算法,用于将二维图像中的建筑物分割为分离的墙面集合,可满足路标识别、导航、场景理解的实时性需求。该算法利用从图像中提取的线段和消隐点信息,将高维的建筑墙面分割问题转化为一维分割问题。实验结果表明,该方法具有精确度高、鲁棒性好、效率高等优点。

Facade Segmentation Based on Lines and Vanishing Points

Segmenting facades from buildings is important to a variety of vision applications operating in outdoor urban environments which include image-based modeling,landmark recognition,autonomous navigation and urban scene understanding.The method based on Structure-from-Motion relies on multiple images,and the cost of computation is high,and the method based on prior knowledge about arbitrarily shaped repetitive regions cannot handle non-repetitive facades segmentation.In order to segment facades simply,fast and accurately,an automatic algorithm for facade segmentation was proposed,which segmented building to separate facade set.Based on the detected lines and corresponding vanishing points,the facade-segmentation problem was converted to a simple 1D segmentation.The experiment results show that the method is accurate,robust and efficient to facade segmentation problem.