Research On Underground Point Cloud Data Acquisition And Preprocessing System Based On Laser Scanning

The spatial information acquisition of underground mining scene is the foundation of realizing intelligent mining and scene monitoring and of great significance to improve the efficiency of mining.The point cloud data acquisition system based on laser scanning with non-contact,high speed,high precision and strong adaptability has become a hotspot in the field of complex environment 3D modeing and measurement.This thesis designs a 3D point cloud data acquisition and preprocessing system based on the principle of laser scanning,which achieves the 3D spatial information acquisition and has the ability of completing some underground measurement tasks.The point cloud collection device is designed based on the principle of laser scanning and takes a high-speed laser ranging sensor as the core.This thesis designs a scanning structure composed of motor-driving unit,mechanical transmission mechanis mechanism multi-surface prism and high-precision angle encoder,which extends one-dimensional laser ranging to three-dimensional scanning.The point cloud collection device sends data to the host software and receives commands through Ethernet interface.In order to adapt the device to mine environment,this thesis made an explosion-proof design.The host software designed based on Qt framework and Point Cloud Library communicates with the point cloud collection device through standard network protocols,and implements a series of functions such as command transmission and point cloud data reception,storage,visualization,preprocessing,etc.Aiming at the problem that the original point cloud which is acquired by the point cloud collection device contains some outlier and noise point and large amounts of data cause difficulties to post-processing,this thesis thoroughly studies the point cloud preprocessing algorithms such as outlier classification and elimination and surface smoothing and proposes a feature-preserving point cloud compression algorithm.Theeffect of these preprocessing algorithms is verified by experiments.In addition,In order to solve the problem of roadheader's heading window measurement and the space ranging on mining workface,this paper proposes a fast measurement method based on two-dimensional laser scanning and RANSAC algorithm and studies a fast space ranging method via the fixed point scanning of the point cloud collection device.The feasibility of these two methods is proved by experiments.The 3D point cloud data acquisition and preprocessing system that the thesis studies can not only accurately obtain the 3D spatial information of underground mining scene but also completes the task of roadheader 's heading window measurement and the space ranging on mining workface and has the advantages of simple operation and low costs.