Computer Engineering and Applications ›› 2018, Vol. 54 ›› Issue (17): 186-191.DOI: 10.3778/j.issn.1002-8331.1705-0224

Previous Articles     Next Articles

Crack recognition of 3D rock images

XIA Chenmu, TENG Qizhi, QING Linbo, WU Xiaohong, HE Xiaohai   

  1. Institute of Image Information, College of Electronics and Information Engineering, Sichuan University, Chengdu 610065, China
  • Online:2018-09-01 Published:2018-08-30



  1. 四川大学 电子信息学院 图像信息研究所,成都 610065

Abstract: This paper presents a new three-dimensional image crack extraction algorithm. Firstly, surface reconstruction, Laplacian grid smoothing and mesh simplification are performed for each connected component of the three-dimensional rock pore model. According to the triangular mesh area and the grid unit normal vector direction feature, the triangular meshes are divided into different categories. The shape factor is used to determine whether the three-dimensional spatial structure of each triangular mesh class has a crack feature. The morphological dilation operation is performed on the set of voxels contained in the three-dimensional spatial structure with fracture characteristics. The intersection of the dilated voxels set and the voxels set of the connected components of the original three-dimensional rock pore model are rock cracks. The experimental results show that the method has a good effect of crack extraction.

Key words: 3D rock image, crack recognition, K- means clustering, shape factor, morphological dilation

摘要: 提出一种新的岩石三维图像裂缝提取算法。首先对三维岩石孔隙模型的每个连通分量执行表面重建、拉普拉斯网格平滑、网格简化等操作。根据三角网格面积和网格单位法向量方向特征,将三角网格划分为不同类别。利用形状因子判定每个三角网格类构成的三维空间结构是否具有裂缝特征。对具有裂缝特征的三维空间结构所包含的体素点集执行形态学膨胀操作,并与原始三维岩石孔隙模型连通分量的体素点集进行逻辑与操作,与操作结果即岩石裂缝。实验结果表明,该方法具有较好的裂缝提取效果。

关键词: 岩石三维图像, 裂缝识别, K-均值聚类, 形状因子, 形态学膨胀