Max A. Viergever
Image Sciences Institute
University Medical Center Utrecht
Title: Image Registration for Medical Research, Diagnosis and Intervention
The talk will first give a brief overview of our institute's research themes in biomedical imaging.
Next, one of the focal areas in medical image analysis, viz. image registration, will be discussed in more detail. The past few years have shown a tremendous increase in the need of accurate, reliable and fast image registration algorithms. Multimodality imaging requires intrapatient multimodality registration, longitudinal image studies call for - generally nonrigid - intrapatient monomodality registration, dynamic (four-dimensional) imaging requires motion registration, and population studies are in need of interpatient registration methods including atlases for normal anatomy and physiology and for pathology and pathophysiology.
Examples will be given of studies on visualization of brain SPECT/MRI, motion correction of DSA, population analysis of white matter lesions, and image-guided radiotherapy.
Max A. Viergever received the MSc degree in applied mathematics in 1972 and the DSc degree with a thesis on cochlear mechanics in 1980, both from Delft University of Technology. From 1972 to 1988 he was Assistant/Associate Professor of Applied Mathematics at this university.
Since 1988 he has been Professor and Head of the Department of Medical Imaging at Utrecht University, since 1989 Professor of Physics and since 1996 Professor of Computer Science at the same university. Since 1996 he has been Scientific Director of the Image Sciences Institute of the University Medical Center Utrecht, since 1998 Director of the Research School for Biomedical Image Sciences (ImagO), and as of 2002 Manager Education and Research of the Division of Radiology, Radiotherapy, and Nuclear Medicine of the University Medical Center Utrecht.
He is (co)author of more than 400 refereed scientific articles (abstracts excluded) on biophysics and medical imaging, guest editor of eight journal issues, (co)author/editor of 15 books, and has served as supervisor of 80 PhD theses and >100 MSc theses. His research interests comprise all aspects of medical imaging.
Max Viergever is an Honorary Senator of the University of Ljubljana, an Honorary Member of the Dutch Society for Pattern Recognition and Image Processing, and a recipient of the Catharijne Award. He is an Elected Fellow of the Institute of Physics IOP, of the International Association of Pattern Recognition IAPR, and of the Institute of Electrical and Electronics Engineers IEEE.
He is (co)author of two best journal articles (Comput. & Graphics 1996, J. Comput. Assist. Tomogr. 1997) and of eight best conference presentations, and has received two citation awards for most frequently cited articles (IEEE T. Med. Imaging 1988, 1995). He has been a board member of IAPR, IPMI and MICCAI, editor of the Springer book series Computational Imaging and Vision, editor-in-chief of the IEEE Transactions on Medical Imaging, editor of the Journal of Mathematical Imaging and Vision, and has acted as associate editor, guest editor or editorial board member of eleven more international journals.
Laboratory for Multimedia Communication in Telepresence
University of Quebec
�cole de Technologie Sup�rieure
Montreal (Quebec), Canada
Title: DIAR: Advances in Degradation Modeling and Processing
The most up-to-date OCR/ICR algorithms and software are the result of large-scale experiments on the accuracy of OCR systems and the proper selection of the size and distribution of training sets. Degradation models are the key factor in the improvement of OCR technology, and they represent the state-of-the-art for conventional printed materials. In recent years, new imaging media, new definitions of textual information, and the need to process low quality document images present new challenges to the degradation modeling community. Some of these challenges, as well as future research directions, will be discussed, as will well-developed degradation models. In particular, the restoration and enhancement of degraded single-sided and multi-sided document images which have suffered bleed-through or shadow-through will be targeted.
This paper will be authored by Mohamed Cheriet and Reza Farrahi Moghaddam.
Mohamed Cheriet was born in Algiers (Algeria) in 1960. He received his B.Eng. from USTHB University (Algiers) in 1984 and his M.Sc. and Ph.D. degrees in Computer Science from the University of Pierre et Marie Curie (Paris VI) in 1985 and 1988 respectively. Since 1992, he has been a professor in the Automation Engineering department at the �cole de Technologie Sup�rieure (University of Quebec), Montreal, and was appointed full professor there in 1998. He co-founded the Laboratory for Imagery, Vision and Artificial Intelligence (LIVIA) at the University of Quebec, and was its director from 2000 to 2006. He also founded the SYNCHROMEDIA Consortium (Multimedia Communication in Telepresence) there, and has been its director since 1998. His interests include document image analysis, OCR, mathematical models for image processing, pattern classification models and learning algorithms, as well as perception in computer vision. Dr. Cheriet has published more than 200 technical papers in the field, and has served as chair or co-chair of the following international conferences: VI'1998, VI'2000, IWFHR'2002, and ICFHR'2008. He currently serves on the editorial board and is associate editor of several international journals: IJPRAI, IJDAR, and Pattern Recognition. He co-authored a book entitled, "Character Recognition Systems: A guide for Students and Practitioners," John Wiley and Sons, Spring 2007. Dr. Cheriet is a senior member of the IEEE and the chapter chair of IEEE Montreal Computational Intelligent Systems (CIS).
Instituto Superior T�cnico
Instituto de Telecomunica��es
Title: Analysis for Video Coding: Moving it from the Encoder to the Decoder
Though without noticing, a growing percentage of the world population uses presently image, video and audio coding technologies on a rather regular basis. These technologies are behind the success and quick deployment of services and products such as digital pictures, digital television, DVDs, and MP3 players. The main purpose of digital audiovisual coding technologies is to compress the original digital information into a much smaller number of bits, without an unacceptable subjective degradation of the signal quality.
The current most used video coding paradigm is based on temporal prediction between video frames to exploit the temporal redundancy, transform coding (e.g. using DCT, Discrete Cosine Transform) to exploit the spatial redundancy, quantization of the transform coefficients to exploit irrelevancy (related to the human visual system limitations), and entropy coding to exploit the statistical redundancy of the created coded symbols. To achieve high compression efficiency using the predictive video coding paradigm, encoder analysis plays a major role for motion estimation, coding mode decision, coefficients selection, quantization step choice, rate control, etc. The better is the encoder analysis, the higher the compression efficiency, typically at the cost of some encoder complexity. The evolution of standard video codecs has stressed this effect by increasing the number of coding modes, the number of prediction frames, the number of block sizes for motion estimation, etc. always asking for a more sophisticated and complex encoder analysis to achieve higher compression efficiency. This coding paradigm, where the encoder is the �analysis center�, fits especially well some applications, such as broadcasting, using the so-called down-link model since few encoders typically provide coded content for millions of decoders.
Around 2002, a video coding paradigm has emerged which has been changing the role, and location of analysis for coding tools. At the light of the Slepian-Wolf theorem from the 70s, the minimum rate needed to independently encode two statistically dependent discrete random sequences, X and Y, is the same as for joint encoding. Later, the Wyner-Ziv theorem has stated that, under some hypothesis on the joint X and Y statistics, when the side information (i.e. the correlated source Y) is made available only at the decoder there is no coding efficiency loss in encoding X, with respect to the case when joint encoding of X and Y is performed even if the coding process is lossy (and not lossless anymore as for the Slepian-Wolf case). The two theorems suggest that it is possible to independently encode two statistically dependent signals and jointly decoding them, while approaching the coding efficiency of conventional predictive coding schemes, which rely on joint encoding and decoding instead. The new video coding paradigm, known as Wyner-Ziv (WZ) video coding avoids the computationally intensive temporal prediction loop at the encoder, by shifting the complex process of exploiting the temporal redundancy to the decoder. Again, the better is the decoder analysis, the higher the compression efficiency, typically at the cost of some decoder complexity. This new coding paradigm, where the decoder becomes the novel �analysis center� is especially relevant for a large range of emerging application scenarios, using the so-called up-link model, such as video surveillance, and visual sensor networks.
This talk will present and compare the conventional, encoder prediction based video coding paradigm versus the novel, decoder estimation based video coding paradigm, especially from a video analysis perspective. Besides the basic concepts and tools, the novel benefits, problems and challenges will be presented in the context of a specific Wyner-Ziv video coding architecture.
Fernando Pereira is currently Professor at the Electrical and Computers Engineering Department of Instituto Superior T�cnico. He is responsible for the participation of IST in many national and international research projects. He acts often as project evaluator and auditor for various organizations. He is an Area Editor of the Signal Processing: Image Communication Journal and is or has been an Associate Editor of IEEE Transactions of Circuits and Systems for Video Technology, IEEE Transactions on Image Processing, IEEE Transactions on Multimedia, and IEEE Signal Processing Magazine. He is a Member of the IEEE Signal Processing Society Image and Multiple Dimensional Signal Processing Technical Committee and of the IEEE Signal Processing Society Multimedia Signal Processing Technical Committee. He was an IEEE Distinguished Lecturer in 2005 and IEEE Fellow in 2008. He has been a member of the Scientific and Program Committees of many international conferences and has contributed more than 200 papers. He has been participating in the work of ISO/MPEG for many years, notably as the head of the Portuguese delegation, Chairman of the MPEG Requirements Group, and chairing many Ad Hoc Groups related to the MPEG-4 and MPEG-7 standards. His areas of interest are video analysis, processing, coding and description, and interactive multimedia services.