Artificial
vision systems based on early-cognitive cortical processing (ECOVISION)
IST – 2001 - 32114
Funding period:
2002-2005
URL:
http://www.pspc.dibe.unige.it/ecovision/
Silvio.P. SABATINI
Department of Biophysical and Electronic Engineering
University of Genoa
Via Opera Pia, 11a
I-16145 Genova, ITALY
tel: (+39) 10 353 2092
fax: (+39) 10 353 2289
e-mail: silvio@dibe.unige.it
Participants
KU Leuven, Prof. M. v. Hulle
Univ. Münster, Prof. M. Lappe
Universita' degli Studi di Genova, Prof. S.
Sabatini & Prof. G. Bisio
Univ. de Granada, Prof. E. Ros
HELLA Hueck KG, Lippstad, Ger, Mr. M.
Mühlenberg
Goals of ECOVISION:
The
goal of this co-operative project is to investigate and implement
advanced
mechanisms that enable the design of high-performance machine vision
systems.
These mechanisms are motivated by early cognitive cortical processing
in the
visual system of vertebrates and part of this project is concerned with
models
of such systems. The central aspect which distinguishes ECOVISION from
other
similar approaches is that we are seeking to employ highly co-operative
multi-modal mechanisms in order to achieve improved image analysis.
This
becomes possible because, at the same time, we are addressing the
problem of
early visual pre-processing by means of dedicated hardware (FPGAs).
These
front-end algorithms provide in a fast and efficient way the input data
for the
early-cognitive post-processing. The application goal of ECOVISION is
to employ
these techniques in driver assistant systems
Summary of Achievements:
The following algorithms and procedures have
been newly
invented by ECOVISON. All these results are major contributions which
have all
been published (see http://www.pspc.dibe.unige.it/ecovision/pubs/index.html).