Study On The Fractal Theory Based Interpolation On Data Of DEM And 3D Display Technique

The fractal interpolation based on the self-similar of data interpolates data by certain mathematical model. The fractional Brownian motion (fBm) is one of. the most effective models that are applicable for describing the fractional characteristic of terrain surface. The advantage of this model is that it can control the shape of surface with few parameters and reconstruct the realistic terrain surface.In the process of interpolating, the calculation of fractional dimension is the most important problem of all. It must calculate fractional dimension within certain range because the self-similar presents in certain scale range. Furthermore, operating against the difference of entire and local data, it needs to divide the data and calculate their fractional dimensions distinguishingly.On basis of fractal theory and the principle of fractional Brownian motion, this paper puts emphasis on the following parts: Firstly, the determination of fractal non-scale range on calculate fractional dimension is discussed, which adopts method associated artificial cognition with track decision. Secondly, primeval data are segmented into blocks to evaluate their sub-fractional dimensions. In this process, the maximal value of sub-fractional dimensions in different directions is prioritized. Thirdly, to improve traditional method (midpoint displacement algorithm), various coefficient scales are used in interpolating data.To the end of this paper, a series of experiments on interpolating data are conducted, which indicate that theinterpolation improves the precision of results and avoids the creasing problem effectively. Finally, a 3-D simulated display about results is accomplished by OpenGL technique.