TC18 "Discrete Geometry" Newsletter (February 2010)


1. Preliminary Information on DGCI 2011

2. Workshop on Applications of Discrete Geometry and Mathematical Morphology,
   August 22, Instanbul, Turkey

3. Short Report on DGCI 2009


1. Preliminary Information on DGCI 2011

Hi everybody,

at DGCI in Montreal, the decision about location and dates of the next DGCI has been made. According to the France/Rest-of-World alternation principle, DGCI 2011 is France's turn, and you have probably already found out that the next DGCI will take place on

    April 6-8, 2011 in Nancy, France

The conference will be organized by the ADAGIo team of the LORIA laboratory under the guidance of Isabelle Debled-Rennesson. She's already very active inviting interesting keynote speakers :-)
According to current planning, papers are due on

    September 6, 2010

where topics of interest include (but are not limited to):

    * Models for Discrete Geometry(grids, discrete objects,
      discrete model properties, digitization schemes, metrics,...)
    * Discrete and Combinatorial Topology
    * Geometric Transforms (image mappings, distance transformation,
      skeletons and medial axis,...)
    * Discrete Shape Representation, Recognition and Analysis
    * Discrete Tomography
    * Morphological Analysis
    * Discrete Modelling and Visualization
    * Discrete and Combinatorial Tools for Image Segmentation and Analysis

You can find full details about the conference on the web site

2. Workshop on Applications of Discrete Geometry and Mathematical Morphology

After DGCI 2009, Annick Montanvert came up with an interesting new idea: Wouldn't it be useful to have a forum were experts of discrete geometry and mathematical morphology could share their knowledge with users and get feedback about application experience and problems?

In order to realize this idea, we (Annick Montanvert, Ullrich Koethe, and Piere Soille) applied for a ICPR pre-conference workshop, and our proposal was accepted. So, the "Workshop on Applications of Discrete Geometry and Mathematical Morphology" (WADGMM 2010) will take place on

    August 22, 2010 in Istanbul, Turkey

The workshop is specifically designed to promote interchange and collaboration between experts in discrete geometry/mathematical morphology and potential users of these methods from other fields of image analysis and pattern recognition. It will comprise four invited review talks by leading researchers in the field, followed by sessions promoting direct exchange and discussion, supported by posters and software demos. We seek posters and software demos that highlight important practical innovations, including

    * Efficient and theoretically proven representations for image segmentation
    * Fast and accurate algorithms to extract topological and geometric features
    * Connections between discrete geometry/mathematical morphology and
      other fields of pattern recognition (optimization, statistics, machine
      learning etc)
    * Parallelization (cluster-based, GPU-based)
    * Application to higher-dimensional data with emphasis on biomedical and
      remote sensing (3D, multi-temporal, multi-sensor, multi/hyper-spectral
    * Thorough comparison with alternative approaches in interesting application

Papers are due on

    April 10, 2010

You can find full details about the workshop and the submission procedure on the web site

3. Short Report on DGCI 2009

As you all know, the 15th Conference on Discrete Geometry for Computer Imagery (DGCI 2009) took place in Montreal, Canada from September 30 to October 2, 2009. It was organized by Srecko Brlek, Xavier Provencal and their team in the nice conference hall of the Université du Québec à Montréal. Many thanks to the organizers for a very enjoyable conference!

In total, 86 participants from all over the world attended and listened to 20 oral and 22 poster presentations. The proceedings were published by Springer as LNCS volume 5810 (see

Besides the usual topics of discrete geometry, a relatively high proportion of the presentations concentrated on combinatorics/pavings/word theory. Keynote talks were given by Valérie Berthé (Montpellier, France) on arithmetic discrete planes, Anders Kock (Aarhus, Denmark) on synthetic differential geometry, and Pierre Gauthier (Montreal, Canada) on climate modelling. All in all, this gave the conference a somewhat theoretical flavour. We must be careful in future DGCI editions to keep a balance between theory and applications, and to emphasize the "Imagery" side of our research. We should strive for results that have a direct impact on challenging image analysis problems, e.g. fast and accurate 3-D algorithms for large volume data (gigavoxels are getting common nowadays!) and experimental comparisons with alternative paradigms (such as machine learning or optimization). This view is a major motivation behind the above mentioned WADGMM 2010 workshop which is explicitly meant to reduce the gap between theory and practice. Another good suggestion has been to work toward a closer collaboration with the mathematical morphology community. I'd like to hear your opinions about these issues.

A new event at DGCI was the first installment of a "Best Student Paper Award". The jury had a tough job selecting the best among several very good candidates. The prize was won by the paper

    * Alexandre Dupas, Guillaume Damiand, Jacques-Olivier Lachaud:
      Multi-Label Simple Points Definition for 3D Images Digital Deformable

but there were many interesting papers and the competition was very challenging.
We think that this prize should become a regular event at DGCI conferences, as it will help attracting promising young researchers to our field and promotes their work.

A more detailed report on the conference by Srecko and Xavier can be found at

Best regards
Ulli (with Joost and Guillaume)