关键词: 露天矿, 台阶线, 激光雷达, 数量场梯度, 最小二乘法

Abstract: In order to efficiently analyze the terrain characteristics of open-pit mine, using UAV LiDAR to scan mine, this paper proposes an automatic extraction algorithm of step lines from point cloud in open-pit mine based on gradient of scalar field,which is referred to as scalar field gradient method. Firstly, the elevation information in point cloud is used as the scalar field corresponding to each point. Then calculate modulus of gradient of scalar field at each point, after analysis the ranges of modulus of gradient of the road and slope scenes are [0.00, 0.03] and [0.28, 0.35]; and then screen out the feature points of step lines according to the range of modulus of gradient.Finally, the feature points are indexed by a quadtree, and least-squares fit straight line in space for each leaf node, and fitted step lines is output. By comparison with step lines extracted by existing software,scalar field gradient method proposed in this paper has the advantage of less manual intervention, and the error rate of extracted step lines can be lower than 10%, and the position accuracy is better than 0.6 meters. The research results can greatly improve the automatic generation efficiency of open-pit mine surveying and mapping products, and have a good application prospect.

Key words: open-pit mine, step line, LiDAR, gradient of scalar field, least square method