摘要
本文介绍了从大脑模型数据点拟合构造出大脑数学模型的方法。模型数据点是通过CT 断层图象得到,在拟合过程中采用了所谓自适应细分算法。其过程是:开始用少量数据点构造出一个粗略数学模型,然后再根据数据点自动地逐步修改不满足给定逼近指标的区域,直到收敛。曲面的表示采用具有 G^1几何连续性的分片参数双三次 Bernstein-Berzier曲面。该算法的优点是对不满足指标的区域修改基本上是局部的,大大压缩了数据量,减少了计算要求,拟合效果良好。
This paper provides a method for the 3-D brain modelling throughsampled data.The sampled data is obtained from the OT images of plaster statueof human brain.The method is based on an adaptive subdivision approach whichbegins with a rough approximating mathematics model and progressively re(?)inesit and correct regions in successive steps.The parametric piecewise bicubicBernstein-Bezier surface is used as surface representation,which possesses Ggeometric continuity.An advantage of the approach is that the refinement isessentially local,so the computational requirements are reduced.Some experi-mental results show that the representation constructed by this method is faithfulto the sampled data.
出处
《计算机应用与软件》
CSCD
1993年第2期13-20,共8页
Computer Applications and Software
关键词
曲面拟合
大脑
数学模型
CAD
建模
Modelling
adaptive subdivision
surface fitting
3-D display
