Information Based Services and Environments

1
Information Based Services and Environments
S. DElia EOP Service Support and Ground Segment
Technology ESA ESRIN, via G. Galilei, Frascati,
Italy
2
Table of Contents

• KIM
• Information Based Services
• KEO
• Conclusions

3
Knowledge-based Information Mining

• KIM permits to interactively search large or well
  characterised areas in all images of a collection
• Training can be reused
• Output
• Image identifiers
• Feature maps / GIS objects

http//kes.esrin.esa.int/kes
4
KIM Primitive Features
5
KIM Test Collections
6
Table of Contents

• KIM
• Information Based Services
• KEO
• Conclusions

7
IBS MERIS Cloud-free Subscription
SSE Service Support Environment KIM
Knowledge-based Information Mining MIMS MERIS
Information Mining Services
8
IBS MERIS Cloud-free Subscription

• Automatically informs about availability of new
  MERIS L1 RR products, with cloud coverage lt user
  defined threshold within user area of interest
• Permits product downloading by clicking on
  provided product identifier

9
IBS Cloud-free Suite

• MERIS Cloud-free Subscription service
• Available to ESA Investigators (includes ESA
  product download)
• Forthcoming MERIS Cloud-free services (search
  which past images are free from clouds over area
  of interest)
• Investigators can select and download archived
  products
• Other users can select catalogue entries
• Cloud-free services being implemented also for
• ALOS
• MODIS
• AVHRR
• Others (on request)

10
IBS Urban Class - Landsat CAT-1

• Automatically classifies ESA Landsat images,
  providing a GeoTIFF map with
• Sealed areas (white) Unsealed areas (black)
  Water surfaces (grey)
• Based on Neural Network algorithm of Tor Vergata
  University (UTV)- EO Lab.
• For Investigators on ESA Landsat collection
• UTV version for all users classifies a user
  provided image

Original (RGB 431)
Classified
Amsterdam
11
IBS Spectral Classes - Landsat CAT-1

• Applies the automatic MEEO (http//www.meeo.it/)
  SOIL MAPPER classification algorithm to Landsat
  TM images of ESA collection over Europe
• Chains catalogue search, image access,
  radiometric correction, classification,
  reformatting

Landsat 7 ETM (RGB 541) of Vicenza (Italy) in
the Po valley close to the Alpine area
SOIL MAPPER classification map, with 72 output
spectral categories in pseudo colours
12
IBS Spectral Classes - Landsat CAT-1

• Identifies several spectral categories belonging
  to classes
• Vegetation
• Rangeland
• Bare soil and Built-up areas
• Water
• Clouds
• Snow and Ice
• Identifies single spectralcategories for classes
  
• Pit bogs
• Greenhouses
• Unknowns
• Available only to ESA Investigators
• MMEO version exists for all users on user
  provided image

Lipari Island
Landsat 7 ETM (RGB 541)
SOIL MAPPER classification map with 72 spectral
categories
13
IBS SOIL MAPPER Extensions

• Classification forthcoming for
• Spot 4 / 5
• MODIS
• AVHRR
• A(A)TSR
• Demonstration on MODIS forthcoming for
• Fire detection
• Smoke identification
• Burned areas identification
• Burned biomass estimation
• All SOIL MAPPER based services are free-of-charge
  for
• Investigators against ESA collections
• All users until end of 2008 on their own images

14
IBS Other Services

• MERIS Global Vegetation Index (MGVI) Subscription
  Service
• Regional maps of vigour and amount of vegetation
• Select calculation time window and time interval
  between deliveries
• Delivery via Data Dissemination System
• Multi Source Fire Detection (forthcoming)
• Geographical co-ordinates of day / night forest
  fires in quasi real time
• Ground resolution 3 km minimum area extension
  0.3 ha
• Test / validation starting over Senegal
• Multi Source Fire Monitoring (forthcoming)
• Integrated picture of past forest fires for user
  defined time window
• Co-ordinates, extension, temperature, confidence
  level from Fire Detection
• Monitor forest fire development for statistics,
  environment / risk management

15
Table of Contents

• KIM
• Information Based Services
• KEO
• Conclusions

16
KEO Positioning
Service Provider / Consumer
Web Services
Non-EO Domain Services (GIS)
SSE
EO Domain Services
EO Ground Seg. Services

• Catalogue
• Order
• Delivery

• Plan
• Process
• Archive

17
Knowledge-centred Earth Observation

• KEO will be a distributed component-based
  processing environment, to
• Create and semantically identify internal or
  external Processing Components
• Create Processing Components also from
  interactive training (no code writing)
• Chain graphically Processing Components into
  Processing Chains
• Create and easily publish chainable SSE services
  from Processing Chains
• Provide SSE access to extracted information
  (stored onto OGC servers)
• Use Data Components
• KEO will provide an environment for Reference
  Data Sets
• Collections of heterogeneous data and
  information(images, documents, DEMs, GTCs,
  photos, processors, etc.)
• Growing with external contributions
• ESA RDSs (2008) Classification, Time-series,
  Ortho-rectification, Polar

18
KEO Concept
KEO
19
Graphical Designer
20
IIM Time Series

• This project addresses
• generic methods and specific algorithms
• for detection of spatial and/or temporal
  evolutions
• in regularly / irregularly sampled image time
  series
• from SAR or optical sensors
• Algorithms will be available in KEO

21
Table of Contents

• KIM
• Information Based Services
• KEO
• Conclusions

22
SSE Summary

• SSE, open e-Collaboration environment for Service
  Providers
• Extends EO GS interface into service provision
  (catalogue, order, dissemination)
• Permits easier, combined exploitation of EO and
  other domains data
• Makes easier prototyping and demonstration of
  services
• Supports service chaining for the creation of
  complex services
• Provides high visibility of services through a
  one stop shop
• SSE (or concept) used by interoperability
  demonstrators
• By many European projects EOLES, InterRisk,
  Orchestra, Sany, Win
• Outside Europe (CSE in Senegal)
• For European convergence on EO Ground Segments
  (HMA suite)
• In related implementations by ESA (DAIL) and
  others (e.g. CNES for Pleiades)
• For defining interfaces with other domains
  (INSPIRE)
• For international test-beds (e.g. FedEO for GEO)
• For possible worldwide partnerships (GEOSS)

23
Possible SSE Network
24
SSE Use

• SSE Technology
• Unique in the EO domain (similar implementations
  in B2B)
• Supports / inspires many European projects
• Fosters increasing demand for orchestration (risk
  of incompatible developments)
• Can support a network of service portals (local,
  regional, thematic, global)
• Is suitable also for research / science
• SSE Infrastructure can support SPs needs (EOVox
  Workshop)
• Directories of Services and Providers for easier
  cooperation
• Access to services for chaining (partnership, no
  redevelopments)
• Cheap methods to reuse the one-off processes,
  set-up in research mode
• SSE extensions ongoing/planned towards In-Situ,
  Grid, Ontology / Semantics

25
SSE Characteristics

• No ESA logo on Portal / Services
• Service Provider can personalise own services
• Neutral ESA moderation
• Common service discovery / access point
• Similar look-and-feel across services for user
  convenience
• One Service Provider responsible vs. users for
  chained services
• Service chaining also within a single Service
  Provider
• Can bridge EO and GIS worlds

26
SSE Support Areas
Researchers

• Data Providers

Value Adders
Service Providers
End Users
27
Your Cooperation / Feedback

• ESA going to set-up a Focus Group to drive SSE
  evolution / adaptation (possibly through EARSC
  participation)
• ESA ready to provide as possible Information
  Based Services or higher-level products to
  simplify further service provision
• Round tables today meant to get your feedback on
• Environment (positive / negative aspects)
• New requirements / missing functions
• Operations
• Service creation / provision limiting factors
• External influences (e.g. data policy)
• Conditions for sustainability of your service
• Needed Information Based Services or higher-level
  products
• others

Thank you