室内3D点云模型的门窗检测  

Door and Window Detection in 3D Point Cloud of Indoor Scenes

作　　者：沈乐 李桂清[2] 冼楚华[2] 江洋 熊赟晖 Shen Le;Li Guiqing;Xian Chuhua;Jiang Yang;Xiong Yunhui(School of Mathematics,South China University of Technology,Guangzhou 510640;School of Computer Science & Engineering,South China University of Technology,Guangzhou 510006)

机构地区：[1]华南理工大学数学学院,广州510640 [2]华南理工大学计算机科学与工程学院,广州510006 [3]School of Computing Science and Digital Media,Robert Gordon University,Aberdeen AB107QB

出　　处：《计算机辅助设计与图形学学报》2019年第9期1494-1501,共8页Journal of Computer-Aided Design & Computer Graphics

基　　金：国家自然科学基金(61572202);广东省自然科学基金(S2013020012795,2017A030313347)

摘　　要：为了检测室内3D场景中的门窗信息,提出一种3D-2D-3D的门窗检测算法.首先在3D室内场景点云模型中多角度旋转拍照,获取点云的2D图像;然后对2D图像进行门窗目标的粗检测,得到门窗在图像中的大致范围,并将此2D信息返回到3D点云数据中,得到包含门窗的局部点云数据;最后提取局部点云数据的轮廓线及其交点,通过优化得到门窗特征角点的位置信息.实验结果表明,采用这种“整体-局部”策略的算法能有效地检测出3D室内场景中门窗的位置信息.This paper proposes a 3D-2D-3D algorithm for doors and windows detection in 3D indoor environment of point cloud data.Firstly,by setting up a virtual camera in the middle of this 3D environment,a set of pictures are taken from different angles by rotating the camera,so that corresponding 2D images can be generated.Next,these images are used to detect and identify the positions of doors and windows in the space.To obtain point cloud data containing the doors and windows position information,the 2D information are then mapped back to the origin 3D point cloud environment.Finally,by processing the contour lines and crossing points,the features of doors and windows through the position information are optimized.The experimental results show that this“global-local”approach is efficient when detecting and identifying the location of doors and windows in 3D point cloud environment.

关 键 词：室内场景 点云 目标检测 门窗检测 特征角点 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]