Welcome to NUMIP homepage
The research activities of the SIGNAL PROCESSING
and TELECOMUNICATION GROUP at DIBE are summarized in the following
sections.The research of the Numip area is focused on image processing
and analysis techniques at a numerical level, rather than at a symbolic
level. At present, two approaches are under study. One uses fuzzy
techniques and is centred around a fuzzy intensity connectedness measure,
that can be applied, for instance, to image filtering and segmentation.
The other approach is probabilistic and is implemented in a Markovian
framework for image segmentation and classification.
The group is actively involved in the design of
image processing techniques with the purpose of improving the current
medical practice. They range from segmentation to registration, from
interpretation to machine learning, generally independent from the
particular modality of acquisition (MR, CT, PET, conventional and
acoustic microscopy). A number of national and international contracts
(currently on or just completed) provide a suitable framework to
co-operate with some of the major industries, involved in the production
of medical apparatus, and with famous hospitals all over Europe. The
group is now involved in tele-medicine and remote consultation and
processing of medical images.
Remote-sensing (RS) applications
In the recent past, the area has been involved
more and more in the remote-sensing domain. The enormous amount of RS
data calls for automated and robust image analysis techniques. In 1996, the
area started a project for the Analysis and segmentation of
remote-sensing images for land-cover mapping.
Contact Person: Prof. Silvana Dellepiane
Department of Biophysical and Electronic Engineering
University of Genova, Italy