基于三维激光扫描技术的岩体结构面智能识别与信息提取

Intelligent identification and extraction of geometric properties of rock discontinuities based on terrestrial laser scanning

岩体结构面几何参数是评价岩体稳定性与渗流特征的重要研究基础。采用三维激光扫描技术,获取岩体露头点云数据,利用改进的区域生长法与解析几何理论,实现了岩体结构面智能识别与信息提取。研究结论如下:(1)通过对区域生长法进行改进,引入节点法向量判别指标,设置平整性检测阈值_1与区域生长阈值_2,可以有效地进行岩体结构面的智能识别。基于提出的信息提取算法,获取了识别结构面的产状、间距与尺寸信息;(2)选取不同的平整性检测阈值_1与区域生长阈值_2,研究了阈值对识别结果的影响规律,并提取了阈值选取的一般建议。通过规则平面与公路边坡两实例应用分析,对所提方法的可靠性进行了验证。

The geometric parameters of rock discontinuities are important for stability and permeability analysis of rock mass. The intelligent identification and properties calculation of rock discontinuities based on the improved algorithm of region growing and analytic geometry were made according to the nodded cloud data of exposed rock measured with the terrestrial laser scanning. To identify the rock discontinuities from nodded cloud data effectively,the region growing algorithm was improved by specifying the normal vectors to the nodes as the judgment indicator and setting two thresholds（smoothness detection threshold δ1 and region growing threshold δ2）. The orientation,spacing and scale of rock discontinuities can be computed using the analytic geometry. The effects of two thresholds on the identification results were investigated,and a suggestion for threshold selection was provided according to the difference of normal vectors between the adjacent discontinuities. Additionally,the proposed method was applied to two cases,regular planes and road slopes,to verify the robustness of the algorithm.