Segmentation of unorganized point cloud from terrestrial laser scanner in urban region

Segmentation of unorganized point cloud is the key procedure of 3D model construction. And also the problem of management of massive point clouds can be manipulated with segmentation techniques. Global segmentation parameters and metrics are the most essential and difficulty parts of segmentation algorithm, especially in urban region. An optimization solution for segmentation of unorganized point cloud is presented, which is based on several estimated geometry properties, using fuzzy clustering algorithms, and extracts geometry features and semantic features in urban area. The proposal segmentation method is tested in terms of its validity, applicability with two unorganized point clouds from terrestrial laser scanner.