Font Size: a A A

3D Reconstruction And Display Of Pneumatic Caisson Excavated Surface Based On Laser Scan Point Cloud

Posted on:2009-09-10Degree:MasterType:Thesis
Country:ChinaCandidate:J LiFull Text:PDF
GTID:2178360242995275Subject:Mechanical and electrical engineering
Abstract/Summary:PDF Full Text Request
Pneumatic caisson,one of the widely used underground technologies, is superior in the secondary development of bustling urban underground space,with the tend of building resource-conserving cities.And using laser 3D scanning to monitor the excavated surface in real-time,has a bright prospect in supporting caisson remote excavation.As the current technology applied in actual project can only show the point cloud model, which is difficult to be identified by excavating operators,there is an urgent need to improve the visual level of the surface model.As a monitoring tool,the surface model must show good performances in real-time automatic reconstruction as well as visibility.But because of some poor conditions in caisson work area,such as the limitation of the scanner location and the moving of the equipments,there are many difficulties in reconstruction.This paper presented a 3D reconstruction methods based on laser scanning points cloud,which can be used in building visual model of caisson excavated surface automatically in real-time to support a monitoring function.The paper mainly researched the process of 3D scene reconstruction, such as data acquisition,range registration,integration of scans and meshing,proposed methods in opaque objects extraction,hole-filling and the proper display of known equipments,to rebuild the surface visual model based on process characteristics and reconstruction requirement of caisson.Invalid points caused by opaque object such as equipments can not be distinguished automatically by using common methods in surface laser scan reconstruction.This paper also proposed an algorithm based on scan path to distinguish them by dividing the scan lines into approximate straight sub-lines then using feature extraction.The proposed algorithm can be able to self-adaptive extract the feature of the opaque object,and separate them from surface.A pneumatic caisson excavated surface display system was developed according to the above method,and some assistant function was done for actual demands,such as contour display and depth measurement.The experiments verify that the proposed method can provide a visual and monitoring model in real-time automatically for assassinating caisson remote excavating.
Keywords/Search Tags:automatic pneumatic caisson, excavated surface, 3D reconstruction, point-based data processing, opaque objects feature extraction
PDF Full Text Request
Related items