Special Sessions

The increasing use of computer technology in many areas of economic, scientific and social life is resulting in collections of digital data. In particular, the automatic analysis of images is a challenging and demanding field. With the proliferation of large image databases, arises the problem of the automatic mass analysis of images. The continuation of mass analyzes on the basis of the classical procedures leads to investments of proportions that are not feasible. Therefore, the effective use of large digital data bases requires developments of novel techniques to efficiently analyze mass data of images.

This scope of this special session is to bring together researchers in the image processing, computer vision, pattern recognition and image databases communities. Topics of interest include all aspects of image mass data analysis including, but not limited, the following areas:

• Image segmentation,
• Image retrieval,
• Content-based search techniques,
• Matching and similarity search,
• Image Acquisition procedures for mass data analysis,
• Organization, summarization and browsing of digital image databases
• Applications in specialized domains (multimedia, medical imagery, satellite imagery, biometry).

This special session provides a forum for presenting and discussing research results and applications in the intersection of pattern recognition, image analysis on one side and graph theory on the other side.

In addition, given the avenue of new structural/graphical models and structural criteria (e.g. belief-propagation, specific graphs under the constellation approach, graph-cuts, dominant sets) for solving computer-vision problems like segmentation, clustering, stereo, matching, and object recognition, this special session encourages researchers in this more general context to actively participate in the special session. Furthermore, the application of graphs to pattern recognition problems in other fields like computational chemistry and bioinformatics is also an important topic within the scope of this special session. The list of topics of this special session includes thus (without being limited to):

• graph matching,
• graph based image segmentation,
• irregular (graph) pyramids,
• graph representation of shapes,
• graph learning and clustering,
• graphs vs rigid structures (quadtrees, pyramids),
• graph transformations,
• data mining with graphs,
• belief-propagation methods,
• graph-cuts methods,
• graphs in constellation models,
• graphs in computational chemistry and bioinformatics.

Recent advances in technologic solutions for virtual slide digitization, based on motorized microscopes and high resolution scanners, open new dimensions in histopathology. The ability to observe and analyze whole specimens rather that single microscopic fields of view will deeply modify teaching and analyzing practices in this discipline in a very close future.
Computer-assisted image analysis of virtual slides is able to contribute significantly to the elaboration of new diagnosis and prognosis-assisted tools which will help clinicians in their daily practice. Nevertheless, several problems have to be overcome, due to the huge size of the virtual images, and to the heterogeneity and complexity of the biological cell and tissue samples.

The scope of this special session is to bring together scientists, pathologists, researchers in the field of large image acquisition and processing and also companies involved in the elaboration and distribution of virtual microscopy solutions. Topics which will be discussed concern all aspects of virtual microscopy, including, but not limited to:

• New image acquisition devices
• Compression, storage and visualization of virtual slides
• Content-based analysis techniques for diagnosis and teaching
• Image processing strategies
• Critical analysis of currently available solutions