基于图像的三维裂隙岩体特征提取、简化及网格生成方法

Image based feature extraction,simplification and grid generation of 3D fractured rock mass

为用最少的网格对高分辨率数字图像的三维裂隙特征进行简化表征,提出了一种基于图像的三维裂隙岩体特征提取、简化及其网格生成方法。首先采用中心面点提取算法将裂隙转换为一系列面,并对其进行网格生成、面片交叉标识及网格优化,以生成的裂隙面网格为约束生成四面体网格模型。然后采用体素和形状相似度两个参数对生成的面网格的代表性进行评估,通过最小二面角、半径比、边长比及体积比等网格质量度量参数对生成的四面体网格质量进行了评价。最后,以一个三维二值化岩石裂隙图像数据为例,对所提方法的有效性及可行性进行了验证。结果表明,该方法在保留裂隙主要几何特征的前提下较好地生成裂隙曲面网格。

In order to simplify the three-dimensional fracture features of high-resolution digital images with minimum mesh,an image-based feature extraction,simplification and mesh generation method of three-dimensional fractured rock mass is proposed.Firstly,the central surface point extraction algorithm is used to transform the crack into a series of faces,and the mesh generation,cross identification and mesh optimization are carried out to generate the tetrahedral mesh model with the generated crack surface mesh as the constraint.Then,the representation of the generated surface mesh is evaluated by voxel and shape similarity parameters,and the mesh quality metrics such as minimum dihedral angle,radius ratio,side length ratio and volume ratio are used to evaluate the quality of the generated tetrahedral mesh model.Finally,the validity and feasibility of the proposed method are verified by taking a three-dimensional binary rock fracture image data as an example.The results show that the method can generate the crack surface mesh well on the premise of retaining the main geometric features of the crack.