三维光纤图形特征的体绘制改进方法研究与仿真

Research and Simulation on Improved Method of Volume Rendering of 3D Optical Fiber Image Feature

当前方法采用三维视景渲染的特征体绘制方法,在光纤背景的景深较大的情况下,体绘制对复杂场景的区分度不高,纹理信息的鲁棒性不好。提出一种基于纹理映射和均匀分布网格重构的三维光纤图形特征的体绘制改进方法。在GPU环境下构建三维光纤图形特征的渲染模型,采用矢量边裁剪方法进行特征空间重构,得到三维光纤图形的限定模型边界范围分量,实现对三维光纤图形特征的纹理特征提取预处理,采用网格分割方法进行逐圈扩散,进行均匀分布网格重构,实现体绘制算法改进。仿真结果表明,采用该算法进行三维光纤图像特征体绘制,具有较好的误差收敛性和鲁棒性,绘制精度较高,性能较好。

The traditional method adopts the 3 D render scene features visual rendering method, in greater depth of field optical fiber background, volume rendering in a complex scene, the distinction degree is not high, the robustness of the texture information is not good. A new method of volume rendering based on texture mapping and uniform distribution of mesh reconstruction is proposed. Characteristics of three-dimensional fiber graphics rendering model is constructed in GPU environment, using vector cutting edge method for feature space reconstruction and get the 3D fiber constrained model of boundary components, graphical feature of the 3- D optical fiber texture feature extraction pre- treatment, using the grid segmentation method by circle diffusion, is uniformly distributed and grid reconstruction, realize volume rendering algorithm is improved. The simulation results show that this algorithm is used to draw the 3D optical fiber image feature, which has good convergence and robustness, high accuracy and good performance.