Research And Implementation Of Triangle Mesh Simplification Algorithm

As a new media form, three-dimensional mesh model have gradually involved in people's lives, and have applied in many areas more and more widely, such as virtual reality, animation games and production. Although the level of computer hardware has been rapidly developed, it will bring great pressure to the rendering, transmitting and editing system of computers that the accuracy of these 3D mesh models grows with the development of the precision of data collection and modeling techniques including the total data of the models soared rapidly. One way to solve this problem is to simplify these complex 3D models, which is accordingly reducing the data, accelerating the process speed and sparing storage space by replacing the original complex model by a simple one. Obviously, the simplification of 3D mesh models is a valuable research direction.The three-dimensional terrain is a special three-dimensional object. In recent years, the visualization of three-dimension (3D) terrain has been a hot research topic in Geographic information system (GIS), virtual reality (VR), simulation of battlefield environment and other associated fields, and the compression technology of terrain data also becomes one of the most important research areas. Three-dimensional terrain surface belong to the scope of the representation of the three-dimensional curved surface, therefore, the general simplification method can be also applied to simplify the surface of terrain. However, the terrain description is height-field data, so it is a special class of three-dimensional surface. And the general three-dimensional curved surface description method often fails to fully make use of the characteristics of high-field data. At present, there are many proposed algorithms to simplify terrain, such as real-time optimally adapting meshes (ROAM), a view-dependent multi-resolution terrain simplification based on adaptive quad trees, height field TIN representation based on local entropy et al, and these algorithms all have their own characteristics.In this thesis, the research is about rendering and simplification of terrain models, and an edge contraction simplification algorithm based on DEM terrain is proposed, which is using high-field data of terrain. This algorithm introduces the gradient of triangle, and combines with gradient and the length of edges as weight of the vertex to represent the importance of a vertex, and use the weights to carry out edge contraction simplification. Experimental results show that our algorithm runs fast and is easy to implement.