Data Management of Large 3D Urban Scenes

Data Management of Large 3D Urban Scenes Xavier Lopez Oracle, Inc. USA 3D Data Management Technology Drivers Use Cases for 3D Data Management Overview of Spatial Databases Wrap-up Technology & Business Drivers Massive new sensor hardware capabilities – – Automated Data Capture / Model Creation (sensors) Lidar (point clouds), automated photogrammetry 3D data coming into mainstream – – Mass Market: Consumer-focused systems Benefit from IT scalability, security, and reliability Managing a 3D data enterprise workflow – – Improve performance and scalability of existing workflows Bridging gap between point cloud surveys, GIS, CAD, BIM Files to Databases Challenges: Managing Point Clouds Robust Data Management Challenges: – – – – High Density LIDAR: Sub-meter point spacing (billions of points) Combine with multi-spectral gridded data (terabytes of data) Versioning, Archiving (terabytes, petabytes) Back-up/recovery LIDAR point filtering, visualization, analysis Surfaces and 3D vector models Attribute Data Integration Leverage Grid computing, clustered servers Integrate 3D models into business workflows Associate 3D objects/features to attributes Spatial query across point cloud features Data Transformation – – – – Visualization & Analytics – – – What do Spatial Databases Bring? Scalability: Large seamless 3D scenes: Terabytes Fast Retrieval: Geospatial Indexes on 3D point clouds Partitioning: Manage large seamless scenes Grid computing: Massive Data processing Interoperability: Fuse aerial imagery, close-range airborne, ground video/LIDAR, 2D vector models Spatial analysis: traditional GIS queries on 3D scenes Transactional Updates Enterprise Integration: Integrate 3D models with business information Versioning and Long Transaction Support: Data security, access control, encryption, authent. Open: Support by third party 3D viz and analysis tools Creating Value Added Data Products CAD/BIM server TIN server Point Cloud Repository 3D Model server Image server Spatial DBMS in a Workflow Data Collection Production Dissemination & Exploitation CAD LIDAR OrthoPhotos LIDAR Surveys Photogrammetry Aerial Photos Satellite Imagery COTS Scenes CAD Designs Model/Scene Generation Image Texture Wrapping Versioning Editing/Updates Quality Control Volumetric Analysis 3D Mapping Fly Through 3D analysis Engineering Design Predictive Analysis Navigation Systems Spatial RDBMS 7  Use Cases GIS Analytical Modeling & Simulation Flood Plain Analysis Flood Plain Analysis Petroleum Exploration Petroleum Exploration CAD Infrastructure Design (Super Models) Courtesy Parsons Brinckerhoff Courtesy Parsons Brinckerhoff Google Earth 3D Mash-ups Simulation, Gaming, and VR  Overview: Spatial Databases Customer Requirements Enterprise RDBMS to manage ALL geospatial types – – – – 2D, 3D, Rasters, Networks, Topology, Attributes Native type support, indexing, and analysis Support 2D & 3D coordinate systems support Standards based: SQL, Java, .NET Building City models (Collada, CityGML) Laser scanning (LIDAR, sonar) Geo-engineering (CAD) Surfaces (TINS, DEMS) City Modeling, environmental analysis Real Estate, asset management Personal navigation VR, gaming, simulation Addresses large volumes of 3D point data – – – – Addresses range of 3D application domains – – – – Spatial Database Capabilities Spatial Analysis Spatial Data Types Spatial Indexing All Location/Spatial Data Stored in the Database Spatial DBMS Fast Access to Spatial Data Spatial Access Through SQL   Spatial Databases: Managing all Geodata Types 3D Models (Buildings) Networks (Highway network) Parcels (polygons) Imagery (Satellite) Spatial Data DBMS Structured Networks/Boundaries (persistent topology) Lidar (Point Clouds)  Oracle 11g Spatial 3D Capabilities 3D Functionality in RDBMS Types 3D COORDINATE SYSTEMS Building Models,.. Efficient Storage – SDO_GEOMETRY (3D) Surface – SDO_TIN SDO_POINT_CLOUD Modeling Query Analysis – Scene, Object Modeling 3D Spatial Data Types Simple and composite Solids – – – – – – Solids are composed of closed surfaces It has to have one outer surface and one or more interior surfaces Cube is an example of a simple solid A pyramid is another example of a simple solid Composite solids: formed by multiple solids Always define a single contiguous volume Point Clouds: LIDAR Large volumes of point data acquired by sensors – – LIDAR (Light Detection and Ranging) Seismic sensors Millions of points used to model a scene New data type introduced to efficiently manage this type of point data TIN to create triangulation of such points TINs: Triangulated Irregular Networks Vector-based topological data model represents terrain/surface Contain a network of irregularly spaced triangles 3D surface derived from irregularly spaced points Each sample point has an x, y, z or surface value Node No 1 2 3 4 5 6 . X Y Z 5 3 1 4 6 2 . 6 6 5 4 5 2 . 3 5 6 4 3 2 . Query, Analysis of 3D Data Given a 3D Geometry as Query, Identify data geometries that – – – Intersect the Query (SDO_ANYINTERACT) Within specified distance from Query (SDO_WITHIN_DISTANCE) Nearest to Query (SDO_NN) Relationship: Geometry-Geometry Intersection Length, Area, Volume Analysis Validation of 3D Geometry: Is Solid ‘closed’ ? Extrusion of 2-D Footprint to a 3-D Solid by specifying heights Association of Textures with LabelStrings of Geometry Elements Extraction of Elements using LabelString of Geometry Conversion Functions to/from GML, to KML/Collada, from CityGML Analysis Functions – – – – – – – 3D Coordinate Systems Oracle Spatial 11g supports the following EPSG types Vertical Coordinate Systems: Essentially 1-d coordinate system (w.r.t sea-level etc.) Geocentric: 3-d Cartesian Geographic-2d, Projected-2d: 2-d Ellipsoidal Geographic-3d: 3-d Ellipsoidal Compound Coordinate System Combines A Vertical Coordinate System with Either Geographic-2D or Projected-2D Coordinate System Support transform between different 3D Coordinate Systems Summary Support large seamless 3D scenes: Terabytes Provide bridge to fuse 3D, 2D, CAD, raster data Fusion of aerial imagery, close-range airborne and ground video/LIDAR, with 2D vector models Integrated support for Web delivery Spatial analysis: GIS queries on 3D scenes Transactional Updates Enterprise Integration: Integrate 3D models with business information. Data security, access control, encryption, authent. Open: Support by 3D viz and analysis tools