The Research On Building Of Measureable Three Dimensional Virtual Campus

The flow of this thesis is terra rebuilding to build modeling rebuilding to 3D scene rebuilding and the last programming a execute program, the aim of this thesis is to make an measurable 3D Virtual Reality campus information system executive program of Xian University of Science and Technology.This examination integrates photogrammetry technique, three-dimension model setting-up technique and texture mapping technique. With the help of these techniques, the research has realized the following purposes: 1, the vector processing of aerial photography, make 4D production; 2, the rebuilding of three-dimensional campus landscape based on plant pictures; 3, build the 3D executive program of LinTong campus of Xian University of Science and Technology by integrating and rebuilding of DEM, texture, and build modeling.In the construction of digital city, the building model rebuilding is a main study aspect. Now there is some localization in traditional methods such as Geometric Modeling winch is used in build model rebuilding. The localizations are: 1, it needs the plan shape and the height of building to rebuild the model; 2, lack of reference images in modeling, so it can not guarantee the accuracy of building model, and the error only in the data browser environment can be displayed fully; 3, Artificial collection, artificial establishment of topological relations and artificial editorial modeling ideas, so the entire modeling process laborious and time-consuming. For the shortage of the above, the authors used photogrammetric methods, using photographs taking by ordinary camera to rebuild the building model, and achieved relatively good results.In the construction of digital city, the three-dimensional scene of measurement is another important direction. The author build the exactitude model-building by photogrammetry methods is measurable in the final the digital campus system. However, building models in the system is a whole non-vector data, due to the uncertainty of non-vector data caused the model in system is not accurate than the vector model. this problem need further study.