Volume II-3/W5
ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci., II-3/W5, 531-539, 2015
https://doi.org/10.5194/isprsannals-II-3-W5-531-2015
© Author(s) 2015. This work is distributed under
the Creative Commons Attribution 3.0 License.
ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci., II-3/W5, 531-539, 2015
https://doi.org/10.5194/isprsannals-II-3-W5-531-2015
© Author(s) 2015. This work is distributed under
the Creative Commons Attribution 3.0 License.

  20 Aug 2015

20 Aug 2015

POINT CLOUD SERVER (PCS) : POINT CLOUDS IN-BASE MANAGEMENT AND PROCESSING

R. Cura, J. Perret, and N. Paparoditis R. Cura et al.
  • Universite Paris-Est, IGN, SRIG, COGIT & MATIS, 73 Avenue de Paris, 94160 Saint Mande, France

Keywords: RDBMS, point cloud, point cloud management, point cloud processing, filtering, indexing, compression, point cloud storage, point cloud I/O, point cloud generalisation, patch, point grouping

Abstract. In addition to the traditional Geographic Information System (GIS) data such as images and vectors, point cloud data has become more available. It is appreciated for its precision and true three-Dimensional (3D) nature. However, managing the point cloud can be difficult due to scaling problems and specificities of this data type. Several methods exist but are usually fairly specialised and solve only one aspect of the management problem. In this work, we propose a complete and efficient point cloud management system based on a database server that works on groups of points rather than individual points. This system is specifically designed to solve all the needs of point cloud users: fast loading, compressed storage, powerful filtering, easy data access and exporting, and integrated processing. Moreover, the system fully integrates metadata (like sensor position) and can conjointly use point clouds with images, vectors, and other point clouds. The system also offers in-base processing for easy prototyping and parallel processing and can scale well. Lastly, the system is built on open source technologies; therefore it can be easily extended and customised. We test the system will several billion points of point clouds from Lidar (aerial and terrestrial ) and stereo-vision. We demonstrate ~ 400 million pts/h loading speed, user-transparent greater than 2 to 4:1 compression ratio, filtering in the approximately 50 ms range, and output of about a million pts/s, along with classical processing, such as object detection.