基于图像处理的小麦叶片形态的三维重建

小麦叶片形态的复杂性使其很难在虚拟场景中真实表现，主要考虑叶片边缘、生长曲线和叶片空间扭曲及卷曲现象，提出一种基于图像的小麦叶片重建方法：采用边缘检测法提取小麦叶片边缘数据，根据小麦叶片生长曲线将边缘数据映射到三维空间，并在此基础上给出了不带约束的叶片模型和带约束的叶片扭曲及卷曲模型。结果表明，该方法采样精度高，便于保持叶片真实形状和实现多细节层次模型(LOD)技术，能较好模拟小麦叶片在空间的扭曲和卷曲现象。

3D reconstruction of wheat lamina shape based on image processing

It is difficult to render the shape of natural wheat lamina in virtual scenes because of its complex and irregular structure. A new image-based 3D reconstruction method, which concentrates on lamina edges, growing curves and the phenomena of lamina twisting and wrapping effect, has been put forward. This method was implemented by employing the edge detection method to extract wheat lamina edge data and mapping the 2D edge data to 3D space according to a wheat curve function. Furthermore, an unrestricted lamina model and a restricted lamina twisting and wrapping model were constructed based on the method. The results show that the new method is convenient for keeping the natural shape of lamina and implementing the Level-of-Detail model, and works well on simulating twisting and wrapping effect of wheat lamina.