Feature Extraction of Laser Scan Data Based on Geometric Properties

详细信息    查看全文

• 作者：Ismail Elkhrachy
• 关键词：Terrestrial laser scanning (TLS) ; Point cloud ; Features extraction ; Geometric properties ; Edge detection ; Segmentation ; Normal vector ; Cultural heritage
• 刊名：Journal of the Indian Society of Remote Sensing
• 出版年：2017
• 出版时间：February 2017
• 年：2017
• 卷：45
• 期：1
• 页码：1-10
• 全文大小：
• 刊物类别：Earth and Environmental Science
• 刊物主题：Earth Sciences, general; Remote Sensing/Photogrammetry;
• 出版者：Springer India
• ISSN：0974-3006
• 卷排序：45

文摘

Terrestrial Laser Scanner (TLS) is an important device to provide three-dimensional geometric information with respect to known reference coordinates system. The data from TLS are often called point clouds due to their high density. They are unstructured and do not have any explicit information about the object, except 3D positional and intensity information to each point. Data segmentation and feature extraction are fundamental problems in the point cloud processing step. These are important prerequisites for object recognition and environment understanding. This paper describes a methodology to use the geometrical information of the scanned object, such as a normal vector to extract and segment object edges. The edge based segmentation technique will be studied and improved. The proposed method is programmed and used to extract different region boundaries for two real TLS data, and the quality of the results shows the effectiveness of the proposed algorithm.