论文部分内容阅读
航空激光扫描(airborne laser scanning,ALS)是一种主动快速获取地球及表面三维信息和反射率信息的数据采集技术,从ALS数据中自动提取地形和建筑物信息是ALS数据处理的重点和难点。借助基于对象分析(object based analysis,OBA)技术,提出建筑物自动提取的流程和相关关键算法,首先利用滤波实现地形信息的提取与非地面点的抽取,缩小搜索范围,同时为建筑物的提取提供地形信息,然后对非地面点进行分割得到地物对象,并借助基于对象分析策略,提取建筑物对象特征并构建特征规则集,实现建筑物对象的自动提取。试验结果表明:提出的算法对建筑物激光点进行自动提取效果较好,提取率和准确率分别达到了90%和75%。
Airborne laser scanning (ALS) is a kind of data acquisition technology that actively acquires 3D information and reflectance information of the earth and the surface automatically. It is the key and difficult point of ALS data to extract the terrain and building information automatically from the ALS data. Based on the object based analysis (OBA) technology, the automatic extraction process of buildings and related key algorithms are proposed. Firstly, the extraction of terrain information and extraction of non-ground points are implemented by filtering, and the search scope is narrowed. At the same time, Then the terrain information is provided. Then, the non-ground point is segmented to obtain the object of the feature. With the aid of the object-based analysis strategy, the feature of the object of the building is extracted and the feature rule set is constructed to automatically extract the object of the building. The experimental results show that the proposed algorithm can automatically extract the laser points in the building, and the extraction rate and accuracy reach 90% and 75% respectively.