摘要
根据矢量建筑图纸的绘制特点,提取具有独立语义信息的块、相邻平行线对、以及封闭轮廓线等特征信息进行建筑图纸识别;通过交互拉伸已识别的矢量数据集和编辑参数化模型库,实现建筑物的三维重建;针对公共设施安全问题研究,设计并生成了适用于火灾模拟和人员疏散模拟的场景语义文件。结果表明,该技术能快速准确地重建出三维建筑模型,其场景语义文件能满足一般公共设施安全问题研究的需求。
On the basis of the characteristics of the vector architectural drawings, extracts feature informations of blocks with unattached semantics, adjacent pairs of parallel lines and closed contour lines for identification. Then, achieves 3D architectural reconstruction through interactive stretching the recognized vector data set and editing the parameterized model database. Finally, aiming at study of the safety of public facilities, designs and generates scene semantic documents applicable to fire simulation and the personnel evacuation simulation. The experiment results show that the method achieves high speed and accuracy in 3D reconstruction of architectural drawings and the generated semantic documents meet the research requirements of the safety of general public facilities.
出处
《湖南工业大学学报》
2009年第4期41-44,共4页
Journal of Hunan University of Technology
关键词
建筑图纸
图纸识别
三维重建
三维场景语义
architectural drawings
drawings to identify
3D reconstruction
3D scene semantics
