Image Information Mining IIM:

1
Image Information Mining (IIM)
Future and Opportunities Klaus Seidel and Mihai
Datcu
2
The future what to do in theory, in technology
and for applications ?
2 0 0 3
Think differently !
3
The future next 3 years
Educate, develop, apply, use Push and valorise
the existing concepts and technology to its
limits to bring into operation the existing
systems for EO applications Prototype new
applications in EO and other fields astronomy,
medicine, biology, Establish formation programs
to enlarge the usage of the technology and
promote education programs to prepare future
development
2 0 0 3
4
The future next 3 years
Operational systems almost ready MERIS, MODIS,
SRTM TerraSAR New applications in medicine,
biology, multimedia. Formation programs Master
on Bio-informatics at Uni. Camerino, Master and
Ph.D. course at ENST Paris, Uni. Salamanca, Uni.
Siegen, TU Munchen.
2 0 0 5
5
The future next 5 years

• Think differently to design new - not yet
  existing - concepts
• Information theory of semantic coding
• Theory of knowledge representation and
  communication
• Multidimensional data and information
  representation
• Theory of information and complexity of massive
  data sets
• Re-investigate unconventional technologies
  optical processing, holography, parallel
  computing, etc.

2 0 0 3
6
The future next 5 years
Research started CNES/DLR/ENST Paris Competence
Center and ESA projects Information theory of
semantic coding Theory of knowledge
representation and communication Multidimensional
data and information representation Theory of
information and complexity of massive data
sets Re-investigate unconventional
technologies optical processing, holography,
parallel computing, etc.
2 0 0 5
7
The future next 15 years

• New theories, new technologies
• Biological and neural information processing
  theory and machines
• Quantum information
• Intuitive and conscious computing
• Intuitive and conscious sensors

2 0 0 3
8
The future next 15 years

• In preparation
• Biological and neural information processing
  theory and machines
• Intuitive and conscious computing
• Quantum information
• Intuitive and conscious sensors

2 0 0 5
9
Difficulties
Too long cycle theory - tehnology users More to
work for inter-domains communication (application
but also theory) More applications on real data
needed
10
Progress
IIM Europe 2x no. of teams involved in the
field of IIM or related areas IIM EO technology
reached the operational maturity IIM, GIS, and
KDD are converging for HR EO images
11
Features
Volume and multimodality of data is growing Data
and information is spatio-temporal and
unstructured Users want to have the
knowledge Interactive is the only way of
operation Exploration is the predominant mode of
interaction Context is critical and
relevant Users are interested in information and
knowledge independent of conjecture
12
Features
Information must be obtained from the
data Databases and search engines were not
designed to provide contents Visualization is
very important HMI are crucial Words signals
semantics Visual perception cognition Memory
latency knowledge - relevance
13
Future
Pre-processing integrated in post processing Data
cleaning integrated with semantic
understanding Non-visual integrated in
visual Latent integrated in intuitive
learning Learning controlled by
unlearning Information integrated with human
centric knowledge Only realtime
14
Methods
Cognitive vs. instruments Reactive systems (may
be more then concious ...) All data must be
availabe (no data, no KDD) Simple, but not naive
Selective, not only addaptive Natural