关键词: 危岩变形, 三维激光扫描, 移动平均法, 点云数据

Abstract: The traditional monitoring method for dangerous rock has some disadvantages, such as difficulty in monitoring point layout and low efficiency. Three-dimensional laser scanning can obtain point cloud data quickly because of non-contact measurement, but data processing is difficult. In this paper, a point cloud data processing method based on moving average method is proposed to improve the deformation recognition rate. Firstly, a new spatial coordinate system is constructed according to the joint characteristics of dangerous rock, so that a certain axis could directly reflect the main slip direction of dangerous rock. Then the feature points in the moving window are calculated by surface fitting, which can effectively reduce the point cloud data. On this basis, the ICP algorithm is used to search the corresponding points. Based on a tunnel project in Wenzhou City, the dangerous rock deformation along potential slip direction in excavation slope was analyzed. The results showed that this method greatly improved the correct rate of deformation identification, and the deformation of the dangerous rock was 10mm in two monitoring times, so certain preventive measures should be taken.

Key words: dangerous rock deformation, three-dimensional laser scanning, moving average method, point cloud data