In this paper, we propose a bottom-up point cloud segmentation method, which utilizes a hierarchical clustering structure combined with the perceptual grouping laws. Our method allows a learning-free and completely automatic but parametric process for segmenting point clouds of 3D outdoor scenes. The experiments using terrestrial laser scanning dataset have demonstrated that our proposed method can achieve good results, especially for complex scenes and nonplanar surfaces of objects. The quantitative comparison between our method and the region growing based method also confirms the effectiveness and efficiency of our method.     «

In this paper, we propose a bottom-up point cloud segmentation method, which utilizes a hierarchical clustering structure combined with the perceptual grouping laws. Our method allows a learning-free and completely automatic but parametric process for segmenting point clouds of 3D outdoor scenes. The experiments using terrestrial laser scanning dataset have demonstrated that our proposed method can achieve good results, especially for complex scenes and nonplanar surfaces of objects. The quantit...     »