基于无人机倾斜摄影的嘉兴老城区住宅识别

为了应对老城区房屋数据信息不足、获取不便的问题，提出利用无人机（UAV）倾斜摄影测量技术实现老城区住宅自动识别的方法. 根据老城区典型多层住宅的外立面特征，确定出阳台构造、立面朝向和房屋长宽比3个参数控制的住宅判别标准. 通过无人机摄影测量获取嘉兴研究区的密集匹配点云、数字正射影像（DOM）和数字表面模型（DSM）数据. 融合DOM及DSM数据提取单体建筑轮廓，分割出单体房屋点云. 基于RANSAC算法提取房屋立面点云并确定立面朝向，根据立面的点云空间分布判断立面长度及阳台构造. 试验表明，在研究区应用该方法识别典型住宅的准确率可以达到90%.

Identification of residential buildings in Jiaxing Old City based on unmanned aerial vehicle oblique photogrammetry

A method based on unmanned aerial vehicle (UAV) oblique photogrammetry was proposed to automatically identify residential buildings in old urban areas in order to deal with the problem that the housing data in the old urban areas are incomplete and inconvenient to obtain. A residential buildings discrimination criterion controlled by three features including the existence of balcony structure, the facade orientation and the aspect ratio was determined according to the fa?ade features of the typical muti-storey residential building in the old urban areas. The dense point clouds, digital orthophoto map (DOM) and digital surface model (DSM) were obtained by UAV photography. The DOM and DSM were used to extract the building contour, based on which the point clouds were cut into building-wise ones. The facade point clouds of each building were extracted by RANSAC algorithm, and the orientation of the building was determined. The length of facade and the existence of balcony were determined by analyzing the spatial characteristics of the point clouds. Results show that the correct rate obtained by the proposed method can reach 90% in the study area.