Font Size: a A A

Incremental True Three-dimensional Reconstruction Method

Posted on:2008-12-11Degree:MasterType:Thesis
Country:ChinaCandidate:L D GongFull Text:PDF
GTID:2190360212987842Subject:Cartography and Geographic Information System
Abstract/Summary:PDF Full Text Request
The method of 3-D reconstruction insides or outsides using the distance images obtained by the 3-D laser scanner has gained more and more attentions, it has become the important and advanced subject in the fields of remote sense, mapping and computer science .In my paper, the traditional methods of laser 3-D reconstruction are researched and analyzed , in addition, the scientific items are combined, such as obtaining and using technique of 3-D information which is the BeiJing learning innovation group, and the key techniques of incorporated building and representation of the city 3-D spatial information which is the national science fund item, based on above researches, the incremental 3-D rebuilding method in real scene are putted forward. The contributions in this paper are shown as follows:1.The vectorized method of the 3-D laser scanning is putted forward systematically, 3-D borderline extraction arithmetic based on the distance images is displayed, and finally, the proposition of the invariability of the borderline perspective is proved.2.The convex decomposition arithmetic based on the micro echelon combination are putted forward and the prove of the mathematic provided, so the incremental 3-D model combination are realized.3.The incremental real 3-D reconstruction system are produced, and the complexity of the arithmetic are analysized.4. The incremental real 3-D software are developed, and successfully used in the disposal of the inside or outside of 3-D laser data.The method of the incremental real 3-D reconstruction is not only used in the disposal of the 3-D laser scanning data, it also provides a new orientation and method for the disposal of other large 3-D data.
Keywords/Search Tags:3-D reconstruction, distance images, incremental, plane segmentation, borderline extraction, vectorize, convex decomposition
PDF Full Text Request
Related items