Research On The Simplification Method To The Rendering Model In The Virtual Environment And Its Applications

Virtual assembling technology creates the digital model of the products by making use of the computer to imitate a physical virtual environment, and uses the virtual model to play the role of the solid one, so we are able to simulate and analyze the assembling process, and find out and consummate the interferences or the assembly bugs in time in order to raise the product competition ability in the market. However, there are still many technology difficulties to take advantage of virtual assembly. One of the most difficult is the massive data involved. The massive model data hinders the real-time delivering, for the virtual assembly process contains thousands of equipment and product formation, Therefore, the data reduction method for efficient display is the one of the most essential factor in the application of virtual assembly technologyIn the virtual assembly environment, the boundary condition is very important, but the inside information which completely is not in sight is not cared. According to this features, thesis makes sure the least viewing angles by using the visibility map and spherical algorithm, and then find out all of the possible viewing angles. Then we could reduce the data to render the model faster and more efficently by deleting the inside information which completely is not in sight determined by making use the depth buffer and alpha fuctions of OpenGL.On the other hand, considering the perspective we could render a model of different accuracy levels by utilizing the continuous level of detail which are created based on the QEM algorithm. The detailed models are selected by near observation while the rough models are chosen for real-time display.Finally we develop the simplification effect confirmation system, and compare the simplification effect of the two kinds of algorithms separately and together, and the results demonstrate the two kinds of algorithms could reduce the data effectively and maintain the model details at the same time, while we can reduce more by using them together. Thus, we combine the two types of algorithm at different precisions to implement the continuous level of detail display at the end of the thesis.