动态沉陷区地面激光扫描数据处理关键问题研究

发布时间：2018-06-30 05:20

本文选题：开采沉陷 + 点云　；参考：《中国矿业大学》2013年博士论文

【摘要】：在矿山开采沉陷研究中，为及时了解井下开采对地表的影响程度以便有效地指导地面建、构筑物防护工作或井下生产，通常需要在地表受影响区域内布设观测站进行长期监测。地面三维激光扫描技术引入矿山开采沉陷研究以后，解决了传统观测站数据量不足、观测周期长等问题，但其在该领域中的应用仍然存在许多亟待解决的科学问题。本文结合地面三维激光扫描数据信息量丰富的优点，通过提出或改进各类算法，提出了相应的数据修正模型和多种沉陷信息的提取方法，实现了良好的处理效果。本文主要获得如下成果： 1)针对地面三维激光扫描技术和矿山开采沉陷监测技术的国内外研究现状，详细剖析该技术在矿山开采沉陷监测应用中亟待解决的科学问题，并提出相应的研究内容和技术路线。 2)提出在数据采集前后记录仪器的倾斜状态并取消仪器的自动补偿功能的数据采集方案，并结合地面三维激光扫描原理，对倾斜角度与其测量角度误差之间的几何关系进行推导，得出仪器倾斜对测量角度的误差影响，继而根据三维点云数据中各点坐标与测量角度的数学关系，将各测量角度误差以改正数的形式加到测量角度上进行数据修正，从而消除仪器倾斜对点位坐标获取的影响。 3)利用基于十进制的Morton码对原始点云数据进行直接索引，将坐标与格网一一对应，并通过链表结构对索引后数据进行存储，提高了数据处理的时空效率。继而在网格内若干高程较低点的基础上，采用Kriging插值算法对网格中心点进行插值。继而分别针对DTM和沉陷盆地模型的精度进行了评定，从而检验该数据处理方法的可靠性。 4)针对包含直立树干、电线杆等圆柱体状地物的点云数据，采用点云数据分层特征提取的方法将圆柱体的中心轴予以提取。针对包含房屋等棱角分明地物的点云数据，采用边缘检测算子+Hough变换对点云数据中包含的直线信息进行提取，得到直线间的交点作为特征点予以存储，，并用于后续的水平位移求取，通过与高精度全站仪测得的测站水平位移检测算法的可靠性。 5)针对台阶状裂缝，借助小波变换在信号突变检测方面的优势，对点云数据中各扫描线上的高程突变进行检测，得到裂缝在各扫描线上的相应位置，继而追踪不同扫描线上的突变点，获得台阶状裂缝在点云数据中的位置。针对平面裂缝，通过将点云数据进行图像化处理，继而对点云图像进行加密优化，使裂缝信息在点云图像中得到突显，最后采用图像处理中的边缘检测技术对裂缝信息进行检测和提取，并将裂缝检测效果应用于工程实践中，通过与高精度全站仪测得的裂缝位置进行对比检验了其可靠性。
[Abstract]:In the study of mining subsidence, in order to understand the influence of underground mining on the surface in time so as to effectively guide the surface construction, structure protection or underground production, it is usually necessary to set up observation stations in the affected area of the surface for long-term monitoring. After the introduction of 3D laser scanning technology into the study of mining subsidence, the problems of insufficient data volume and long observation period in traditional observation stations are solved. However, there are still many scientific problems to be solved in the application of laser scanning technology in this field. This paper combines the advantage of abundant information of 3D laser scanning data on the ground, proposes or improves all kinds of algorithms, puts forward the corresponding data correction model and the extraction method of many kinds of subsidence information, and realizes the good processing effect. The main achievements of this paper are as follows: 1) aiming at the research status of 3D laser scanning technology and mining subsidence monitoring technology at home and abroad, This paper analyzes in detail the scientific problems that need to be solved in the application of this technology in the monitoring of mining subsidence. The corresponding research content and technical route are put forward. 2) A data acquisition scheme is proposed to record the tilt of the instrument before and after the data acquisition and cancel the automatic compensation function of the instrument, and combine with the principle of 3D laser scanning on the ground. The geometric relationship between tilt angle and measurement angle error is deduced, and the influence of instrument tilt on measurement angle error is obtained. Then, according to the mathematical relationship between each point coordinate and measurement angle in 3D point cloud data, The errors of each measuring angle are added to the measurement angle to correct the data in order to eliminate the influence of the tilt of the instrument on the acquisition of point coordinates. 3) the Morton code based on decimal is used to directly index the original point cloud data. The coordinates are matched with the grid one by one, and the indexed data is stored through the linked list structure, which improves the space-time efficiency of data processing. Then on the basis of the lower elevation in the grid, the Kriging interpolation algorithm is used to interpolate the center point of the grid. Then the accuracy of DTM and subsidence basin model is evaluated to verify the reliability of the data processing method. 4) the point cloud data containing cylindrical objects such as vertical trunks, utility poles, etc. The center axis of the cylinder is extracted by using the stratified feature extraction method of point cloud data. The edge detection operator Hough transform is used to extract the straight line information in the point cloud data, and the intersection point between the lines is stored as the feature point for the point cloud data including the house and other angular ground objects, and the edge detection operator Hough transform is used to extract the line information contained in the point cloud data. And it is used to calculate the horizontal displacement in follow-up. The reliability of the horizontal displacement detection algorithm obtained by the high precision total station instrument is obtained. 5) aiming at the step crack, the advantage of wavelet transform in detecting the sudden change of signal is obtained. This paper detects the abrupt change of each scan line in the point cloud data, obtains the corresponding position of the crack on each scan line, and then tracks the abrupt point on the different scan line, and obtains the position of the step crack in the point cloud data. In view of the plane crack, the point cloud data is processed by image processing, and then the point cloud image is encrypted and optimized, so that the crack information is highlighted in the point cloud image. Finally, the edge detection technology in image processing is used to detect and extract the crack information, and the effect of crack detection is applied in engineering practice. The reliability of crack detection is compared with the crack location measured by high precision total station.
【学位授予单位】：中国矿业大学
【学位级别】：博士
【学位授予年份】：2013
【分类号】：P642.26;P225.2

【参考文献】