1

AGILE 2012Avignon

25/04/2012

From 2D Maps to 3D Geoservices: Evolving Model of NMAs and

Related Research Issues

Jean-Philippe Lagrange IGN-France & EuroSDR

2

Many changes in mapping / geographic informationData collection

• Highly and often heavy specialised instrumentation

• Which required skilled and trained workers

Data collection mostly limited to national agencies

3

Many changes in mapping / geographic informationData capture

4

• Computerized processes

• Firstly, digital data, vector or raster

Many changes in mapping / geographic informationMap making

5

Many changes in mapping / geographic informationData storage

6

Many changes in mapping / geographic informationData storage

• Cumulated archive until 2007: 50 Tb

• Data archived in 2008 25 Tb 2009 35 Tb

2010 50 Tb2011 100 Tb2012 200 Tb (exp.)

7

Many changes in mapping / geographic informationData distribution

8

Many changes in mapping / geographic informationData distribution

• Until 2007: Only delivery based on physical media (CDs, DVDs, HDs)

• Now growing percentage of delivery based on downloading of data sets

• Starting in summer 2012: More and more data flows (WMS, GML,…)

9

Many changes in mapping / geographic informationBut still a computerised classical process

• Sequential process from stereoplotting to distribution• Well suited for mass production

• Most often layered modelling: Coverages, overlay techniques,…

• Map printing approach

• Data producers issue data to data users• Quality control processes made easier• Issue of fitness for use

10

Many changes in mapping / geographic informationMany more stakeholders

• Local authorities contract out GI collection• Regions: Orthoimagery, high resolution DTMs of specific aeras, road data…• Big cities: High resolution orthoimagery, 3D models, urban specific data

• Decentralised state authorities too• High resolution DTMs

Dozens of data exchange Mous• Lidar data on specific areas (~10-15 / year)• A few orthoimage coverages of départements (~3 / year)→ Growing workload related with quality control

11

Many changes in mapping / geographic informationData creation/update: Volunteered GI

12

A new context for "mapping agencies" Common to most agencies

Still a need for research related with data capture, management, distribution (at large), and data exchange

And a new model which is emerging:• From a sequential and one-way processData capture Map Creation

Data Set consolidation

Distribution

• To loops in a more distributed system

Data capture

Change detection

Data consolidation

VGI

13

Linking National Mapping and Cadastral agencies with Research Institutes and Universities

EuroSDR

• Members of EuroSDR are– National Mapping & Cadastral Agencies,– Research Institutes 2 delegates per country- 1 from academia, - 1 from NMCA

• Delegates from production and research working together

14

EuroSDR approach

In view of optimizing the provision of reference information in a Geoinformation Infrastructure context

15

• Members from Sixteen Countries

• Scientific Commissions• Sensors, Primary data acquisition and Georeferencing• Image Analysis and Information Extraction • Production Systems and Processes• Data Specifications • Network Services

• Two Working Groups• Education• Standards

• INSPIRE Spatial Data Interest Community

EuroSDR

16

• Approx. 15-20 Research Activities at any time

• Official Publications

• Annual Distance e-learning courses– for professional development and up-skilling

• Two general meetings per year (+ Workshops and projects’ meetings)

– Research planning and management– Keynote presentations– Focused discussions –best practice sharing

EuroSDR

17

• Unconventional Platforms for Remote Sensing

• EuroDAC2 – Certification of airborne digital cameras

• Medium Format Digital Cameras - performance

• Radiometric Aspects of Digital Photogrammetric Images

• Mobile Mapping - Road Environment Mapping using

Vehicle-based Laser Scanning

• Towards Integration of Laser Scanning and Photogrammetry

• Image Matching for DSM production

• Radiometric Calibration of Laser Scanning Intensity

•Virtual Globes

EuroSDR

18

• Generalisation

• INSPIRE Atlas of Implementation methods

• Network for GI Service Testing

• Multiple Information Representation And Consistent

Logical Environment – Enabling

• A Persistent Testbed for Geo Web Services - Research

and Education

• Standards (liaison with ISO and OGC)

• Land Use/Land Cover change detection

• 3D Landscape models

EuroSDR

19

• Investigation and monitoring of new sensor systems and platforms

• Related calibration aspects, including digital aerial and satellite sensors, laser and hyper-spectral scanners, SAR sensors, • Unmanned aerial vehicles, and mobile mapping systems

• Geometric data collection and refinement issues • Sensor orientation and geo-referencing, refined image matching, • Digital surface models, digital terrain models• Geometric performance of multiray photogrammetry, LiDAR and airborne SAR

•Radiometric data processing and refinement. •Absolute and relative radiometric data calibration

•Acquisition and update of static and moving objects • Topographic vector data, DTMs and data from other objects• Based on images and and other sources• Merging data of several resolution and accuracy

Key EuroSDR Research topics2011-2014

20

•From 2D cartographic databases to 3D topographic information systems

• Serving the needs of 3D landscape and 3D city models. • All aspects of the whole spatial data chain are to be renewed: spatial modelling, acquisition, management, analysis and visualisation

• Process modelling and interfaces (product diversity / servicing / outsourcing)

• SaaS (Software as a service), PaaS (Platform as a service) and IaaS (Infrastructure as a service)

•Core geoinformation databases • Data modelling, metadata• Currency/maintenance, changes only• Industrial and crowd-sourced strategies for data revision• Privacy enabling

Key EuroSDR Research topics2011-2014

21

• Standards and open software for data and metadata. • Focus on data archives

•Ontologies and translation services• Schema translation• Data integration• Model generalisation, multi-resolution and multi-representation data bases

•Implementation strategies of INSPIRE. • How should INSPIRE 2.0 look like?

• Generalisation, • Cartographic generalisation for traditional and non-traditional displays• 3D generalisation including for mobile (handheld) computing devices

• Delivery and data publishing mechanisms • Geospatial Data Infrastructure and associated topics in Ontologies and translation services• Web-services for dissemination of spatial data

• Data quality• Integration of professionally maintained spatial data meets crowd-sourced data

Key EuroSDR Research topics2011-2014

22

• International Terrestrial Reference Frame & Galileo Terrestrial Reference Frame

• Improvement of vertical measurements

• Better geoid

• Permanent Geodetic Network: Real time GNSS data

Geodetic & Gravimetric Reference FrameworksNeeded for many professional / scientific applications

23

IGN perspectiveData capture• Further automation of DEM / HR DSM calculation

• Automated qualification • Assisted editing

• Web services to edit data• Not only 2D data• But also 3D data

• Exploit the full potential of Lidar data

• Automate the integration of heterogeneous data sets in the context of big data

•e.g., agriculture (CAP data), vineyards, IGN topographic data,…

• New approaches to aerial surveys• UAVs• Light photogrammetric cameras

• Sensor networks with accurate positioning

24

IGN perspectiveVGI : A lot of user specific data

• Great in terms of fitness for use• High potential for faster updates

• Not necessarily well suited in terms of systematic data capture

Combine with reference data or try to duplicate?

25

IGN perspectiveBig Data

• Grid computing - Cloud computing• Common size of computation:

• 2-5 Tb of input data• Or several millions vector features

• Modelling of process flows

• Virtualization of data• 2012: several petabytes (with redundancies)• Issue of interdependencies

• Big data & data analysis

• Distributed architecture and big data• NMAs = reference frame• Other stakeholders= application data, more semantics (user tagging, mashups) & update info

26

• Qualification of data• Geometric: Against reference data?• Semantic: Corroboration? Statistical ‘evidence’? Trustfulness? • Accountability?

• Automated schema matching of qualified data• Ontologies• Related Web Services

• And still:• Generalisation (fully automated, 3D,…)• Space&Time modelling and visualisation

IGN perspectiveAccessing Data & Integrating Data

27

IGN perspectiveCustomisation of Visualisation

28

IGN perspectiveVisualisation with user’s data integrated

From distributed dataFully Automated Map Creation

29

IGN perspectiveVisualisation with user’s data integrated

In 3D (avalanches, urban planning regulations, …)With time dimension (e.g., floods) And for mobile devices…

30Linking Open Data cloud diagram

• Linked data• Not interelated data sets• But GI part of interrelated knowledge chunks

• & Related services (discovery + access)

IGN perspectiveGeographic Information integrated in the Web

F. Scharffe et al, AAAI workshop on semantic cities (2012)

31

• Beyond “NMAs”• Spatial decision / Computer aided spatial reasoning

• Privacy and mechanisms (not only technical) to protect privacy• More and more high resolution data (at the individual level)• Social networks and location • Crowd sourcing

• And many others (see Max Craglia’s paper)

This was only an IGN perspective

32

Thank you for your interest