The Development of a Scientific Permanent Scatterer System

1
ESA-EUSC 2006 Image Information Mining for
Security and Intelligence Mihai Datcu
2
European Image Information Mining Coordination
Group founding members ASI, CNES, CNR, DLR,
EC-IST, ESA, ETHZ, EUSC
3
Archives and Libraries

• Archive a long-term storage area, place or
  collection containing records, documents, or
  other materials of historical interest (thats
  passive and static!)
• Library a depository built to contain books and
  other materials for reading and study (thats
  active and dynamic!)
• What makes the difference?
• Library has
• A catalogue (better then archives) indexing
  books based on multiple criteria, e.g. author,
  title, keywords, domains (ontology!)...
• A librarian one who has the care of a library
  and its contents, selecting the books, documents
  and non-book materials which comprise its
  collection, and providing information and loan
  services to meet the needs of its users

4
Searching Libraries

• Use the catalogue select indexes and search the
  books. Next, read them.
• Walk trough the library browse till you get
  interested...
• A friend told me... go to the material using
  prior information
• Ask the librarian the ideal librarian
• reads all the incoming books
• interprets contents
• associate with other information
• creates categories
• makes inference and understand the inquiry
• dialogues and comments, (communicate) and suggest
  
• does not forget

5
Searching Libraries

• Semantics....
• French Bois Foret
• German Holz Wald

Context
6
Image Semantics
The picture
7
Image Semantics
Librarian Picasso
The picture
Description
8
Image Semantics
Librarian Picasso
The picture
Description
9
Image Semantics
Librarian Picasso
The picture
Description
Librarian Sport fan
Description
10
EO The Challenge The past 30 years EO data
archiving resulted in huge repositories O 1
000 000 multi-sensor images Actually less
than 5 of the data are used The next 10
years new generations of high resolution EO
sensors A double explosion 1. Data
volumes to be acquired 2. Information details
contained in the data
O 1 000 images / day Resolution O 10
cm Optical SAR Multimodal New applications and
users
Pléiades
TerraSAR-X
11
www.coc.enst.fr
12
Information ?(1/resolution)
2007 2007
COSMO 2007
x
13
Information ?(1/resolution)
Landsat 7 11
14
Information ?(1/resolution)
Landsat 7 11
GoogleEarth 110
15
Information ?(1/resolution)
Landsat 7 11
11
11
GoogleEarth 110
16
Information ?(1/resolution)
Intermap
GoogleEarth
E-SAR DLR
17
Image Other Information Knowledge
X-SAR Oct.-April 1993/94
18
The Mind
The ancient Greeks marionettes controlled by
strings (neurone is "string" in Greek).
Descartes mind was seen as hydraulics (fluid
in the supposed nerve tubes). Industrial
Revolution a complex systems of pulleys and
gears. In the 1930's a huge telephone
switchboard. In the 1940's a function of
Boolean logic. In the 1960's a digital
computer. In the 1970's was as hologram. In
2006 no se sabe (nonconosciuto)
19
Intelligence
Gottfredson

• A mind ability to
• reason
• plan
• solve problems
• think abstractly
• comprehend complex ideas
• learn quickly and learn from experience

• Piaget and Bruner
• reasoning
• associative memory
• computational ability
• verbal comprehension
• perceptual speed
• word fluency
• spatial visualisation

20
Image Intelligence

• Images provide
• a wealth of information
• attributes and site detail
• ground-truth characteristics
• The value of images lies also in
• knowing the exact location of acquisition
• spatially referencing
• time referencing

• Visual-Spatial Intelligence
• relationship between line, color, shape, space,
  form
• ability to geometrically manipulate objects
• capacity to create topologic relationships of
  object
• ability to understand the components of visual
  and spatial structures within images

21
Satellite Image Intelligence, but not alone

• Aggregate with
• location-based images or video
• digital maps
• GPS
• background map data
• geospatial intelligence
• text
• human intelligence
• Results in
• decision-making
• documents
• communicable knowledge

22
CONCEPT DE COMMUNICATION AVANCE
Sourcedinformation
Utilisateurs
Extraction de linformation
Représentation sémantique
Inférence du modèle de signal objectif
Inférence du modèle d information subjectif
Modélisation de la conjecture utilisateur
Modèles sémantiques
Modèles syntaxiques
The principle of semantic compositionality the
meaning of a whole is a function of the meanings
of its parts and their mode of syntactic
combination
23
Security
Stimulate to contact sessions chairs to provide
feedback on expectaion from conference topics