Monday June 15th will be a full day of tutorials, one in the morning and one in the afternoon. Registration for tutorials is handled alongside the general SCIA registration process.
The conference organizers are extremely proud to present two excellent tutorial speakers for SCIA 2009.
Tutorial 1: "Principles and Methods of Image Fusion"
Time: Monday, June 15, 09:00 - 12:00
Presenter: Prof Jan Flusser from The Institute of Information Theory and Automation, Praha, Czech Republic
Abstract:
The goal of image fusion (IF) is to integrate complementary multisensor, multitemporal and/or multiview information into one new image containing information the quality of which cannot be achieved otherwise.
This tutorial aims to present a review of recent as well as traditional image fusion methods of various kinds with special emphasis on fusion for restoration and superresolution purposes. The reviewed approaches are classified according to the type of the input images and according to the fusion purpose. Main contributions, advantages and drawbacks of the methods will be discussed in the tutorial. Many practical examples from various application areas (surveillance, medical imaging, remote sensing, robot vision, and astronomy) will be demonstrated. Problematic issues of image fusion and outlook for the future research will be discussed too.
The major goals of the tutorial are
- To provide the researchers with a comprehensive survey of image fusion methods, regardless of particular application areas;
- To present the recent development in the field, particularly in superresolution imaging via fusion.
The target audience of the tutorial are researchers from all application areas who need to integrate and fuse image data of various kind as well as the specialists in image fusion interested in a new development of this dynamic area.
Tutorial 2: "The Dissimilarity Representation for Pattern Recognition:
Introduction and Examples"
Time: Monday, June 15, 13:00 - 16:00
Presenter: Prof. Robert P.W. Duin from Delft University of Technology, Delft, The Netherlands
Abstract:
The dissimilarity representation is an alternative for the use of features in the recognition of real world objects like images, spectra and time-signal. Instead of an absolute characterization of objects by a set of features, the expert is asked to define a measure that estimates the dissimilarity between pairs of objects. As such a measure may also be defined for structural representations such as strings and graphs, the dissimilarity representation is potentially able to bridge structural and statistical pattern recognition.
The tutorial aims to give an introduction of the dissimilarity representation to students and researchers that need pattern recognition techniques in their applications. It will consist of three parts:
- Vectorial representations: features, pixels, dissimilarities. We will explain the problems of features: class overlap, the problems of pixels: overtraining and the potentials of dissimilarities.
- Handling dissimilarity data: the traditional nearest neighbour rule (or template matching) is compared to two alternatives: embedding and the dissimilarity space. This results into two entirely different vector spaces in which classifiers may be trained that may perform much better than the nearest neighbour approach.
- Problems with non-Euclidean data (related to indefinite kernels): in practice many dissimilarity measures used by application experts appear to be non-Euclidean. It will be explained why this is an essential pattern recognition problem. Possible solutions will be discussed.
See further http://ict.ewi.tudelft.nl/~duin/presentations.html#_Representation_1
