一种高斯混合模型组合分类的机载LiDAR城区道路提取方法

城区道路自动提取一直是遥感领域研究的重点和热点之一。针对遥感影像提取易受建筑物和植被遮挡的影响,点云数据提取道路边界又较模糊的不足,提出了一种高斯混合模型组合分类的道路提取方法。该方法利用融合影像即含有色彩信息的点云数据,首先对滤波后点云中的反射强度属性,运用偏度平衡法粗提取道路点云;再对点云数据中的灰度信息和点密度属性采用高斯混合模型组合分类提取道路的种子区域,并利用强度影像扩展和约束该区域;最后运用主动轮廓法和数学形态学方法进一步优化并提取道路中心线。为验证该方法的有效性,分别采取位于国外某城市的两组LiDAR点云数据进行实验。结果表明,该方法可以有效地减弱阴影遮挡对道路提取的影响,提取的道路中心线较为平滑,道路的提取质量达到85%以上。

An Urban Road Extraction Method from Airborne LiDAR Based on Gaussian Mixture Model Combination Classification

The automatic extraction of urban roads has been one of the hotspots in the field of remote sensing.Aiming at the shortcomings that remote sensing image is easily affected by tall buildings and vegetation occlusion and the road boundary extracted from point cloud data is relatively fuzzy,a road extraction method based on Gaussian mixture model combination classification is proposed.This method uses point cloud data that contains color information to fuse the image.Firstly,according to property of reflection intensity of the point cloud after filtering,a skewness balancing method is used to extract the road point cloud.Then,the gray information and properties of point density in point cloud data are classified by the Gaussian mixture model for pattern classification to extract the road seed region,which is expanded and constrained by the intensity image.Finally,the snakes and mathematical morphology are used to extract road centerline efficiently.To verify the effectiveness of the proposed method,two sets of LiDAR point cloud data located in a foreign city are used for experiments.The results show that the proposed method can effectively reduce the impact of shadow occlusion on road extraction,the centerline of the extracted road is relatively smooth,and the extraction quality of the road reaches over 85%.