|
PLENARY SPEAKERS
- Computational Intelligence in Feedback Systems
Prof. Marios Polycarpou
Department of Electrical and Computer Engineering,
University of Cyprus, Cyprus
more info >
- Neural Network Principles for Organisms in Nonstationary Environments
Prof. Daniel S. Levine
University of Texas at Arlington, USA
more info >
- Classification of Infrasound Events: A Neural Network approach
Prof. Fredric M. Ham
Harris Professor of Electrical Engineering
Florida Institute of Technology, USA
more info >
- Image Processing and Image Mining for Biotechnological Inspection Processes
Prof. Petra Perner
Computer Vision and Applied Computer Sciences
Leipzig, Germany
more info >
- Novel Multifusion Imaging Technologies: From Cancer Detection to Semiconductor Wafer Inspection and Nanophotonics
Prof. George Constantine Giakos
Department of Electrical and Computer Engineering
The University of Akron, Auburn Science and Engineering Center
Akron, OH 44325-3904, USA
more info >
- Inverse scattering procedures for active imaging systems at radio frequencies and microwaves
Prof. Matteo Pastorino
Department of Biophysical and Electronic Engineering
University of Genoa, Italy
more info >
- Transcoding MPEG-2 compressed video into H.264/AVC
Dr. Daniele Bagni
Advanced System Technology labs
STMicroelectronics, Agrate Brianza, Italy
more info >
- Frequency Domain Adaptive Filtering in Signal Processing & Communications
Prof. Kosta Berberidis
Computer Engineering & Informatics, School of Engineering,
University of Patras, Greece
more info >
- Knowledge-Based Web Systems for Product Representation and Recommendation
Prof. Bhanu Prasad
Department of Computer and Information Sciences
Florida A&M University
more info >
- Superfast Filters
Prof. Dr. Demetrios G. Lainiotis
more info >
DETAILS
- Computational Intelligence in Feedback Systems
Prof. Marios Polycarpou.
Department of Electrical and Computer Engineering,
University of Cyprus
Abstract
Recent technological advances in computing hardware, communications and real-time software have provided the infrastructure for designing intelligent decision and automated control systems. Based on current trends, high performance feedback systems of the future will require greater autonomy in a number of frontiers. First, they need to be able to deal with greater levels of, possibly, time-varying uncertainty. Second, they need to be able to handle uncertainties in the environment, which will allow the feedback system to be more flexible in dealing with unanticipated events such as faults, obstacles and disturbances. Finally, key advances in distributed and mobile computing will allow for exciting possibilities in distributed decision making and control by agent-type systems. This will require feedback systems to operate in distributed environments with cooperative capabilities. One of the key tools for realizing such advances in the performance and autonomy of feedback systems is "learning." Feedback systems with learning capabilities can potentially help reduce modeling uncertainty on-line, make feedback systems more "intelligent" in the presence of uncertainty in the environment, and initiate design methods for cooperative feedback systems in distributed environments. During the last decade there has been a variety of learning techniques developed for feedback systems, based on structures such as neural networks, fuzzy systems, wavelets, etc. The goal of this presentation is to provide a unifying framework for designing and analyzing feedback systems with learning capabilities. Various on-line approximation techniques and learning algorithms will be presented and illustrated, and directions for future research will be discussed.
About the Plenary Speaker
Marios Polycarpou is a Professor and Interim Chair of the Department of Electrical and Computer Engineering at the University of Cyprus. He received the B.A. degree in Computer Science and the B.Sc. in Electrical Engineering from Rice University, Houston, TX, USA in 1987, and the M.S. and Ph.D. degrees in Electrical Engineering from the University of Southern California, Los Angeles, CA, in 1989 and 1992 respectively. In 1992, he joined the Department of Electrical and Computer Engineering and Computer Science, University of Cincinnati, Cincinnati, Ohio, USA where he reached the rank of Professor. In 2001 he joined the newly established Department of Electrical and Computer Engineering at the University of Cyprus. His teaching and research interests are in computational intelligence, systems and automation, with emphasis on adaptive control, intelligent systems, neural networks, fault diagnosis and cooperative control. Dr. Polycarpou is currently the Editor-in-Chief of the IEEE Transactions on Neural Networks. He has published more than 150 articles in refereed journals, edited books and conference proceedings. He was the recipient of the William H. Middendorf Research Excellence Award at the University of Cincinnati (1997) and was nominated by students for the Professor of the Year award (1996). He is past Associate Editor of the IEEE Transactions on Neural Networks and the IEEE Transactions on Automatic Control, and he served as Vice President, Conferences, of the IEEE Neural Network Society. His research has been funded by a number of agencies, including the European Commission, DARPA, US Air Force, American Water Works Association (AWWA), NASA, Federal Highway Administration (FHWA), ONR, Ohio DOT, and the US Army.
- Computational Intelligence in Feedback Systems
Prof. Daniel S. Levine
University of Texas at Arlington, USA
Abstract
Responses required of biological organisms in nonstationary environments pose complementary challenges for engineering design. For example, vision and other sensory systems must obtain sufficiently accurate representations of external events, yet modify these representations in order to perceive regularities in the environment. Motor control systems must perform the same movement at variable speeds and with variable effectors, yet interrupt a movement when other environmental contingencies need to be addressed. Categorization systems must learn novel events, yet not forget much older events. These systems also must self-organize categories based on regularities in the external world, yet dynamically weight input attributes in response to changes in task requirements.
After sixty years, neural network models inspired by these design issues are approaching the structure and functional organization of the brain. This has led neural network theories such as adaptive resonance to be useful in engineering applications ranging from radar signal detection to robotics to medical and financial data classification. At the same time, these network theories have predicted and been partially confirmed by recent results on the organization of brain areas such as visual and motor areas of the cortex.
About the Plenary Speaker
Daniel S. Levine is Professor of Psychology at the University of Texas at Arlington. Member of Board of Governors, International Neural Network Society (INNS); was President in 1998. Founding member of Executive Committee of Metroplex Institute for Neural Dynamics (MIND), a Dallas-Fort Worth area organization active since 1987 and possibly the oldest neural network organization in the world! He is the organizer of the MIND conferences, which bring together leading neural network researchers from academia and industry. Since 1975, he has written nearly 100 books, articles, and chapters for various audiences interested in neural networks. His research interests include Neural networks; models of frontal lobe function; models of decision making; models of cognitive-emotional interactions; theories of caring and uncaring behavior. He is the Program Chair for International Joint Conference on Neural Networks (IJCNN'05), Montreal August 1-4, 2005. His most known books are:
- D. S. Levine, Introduction to Neural and Cognitive Modeling, Lawrence Erlbaum Associates, 2000
- D. S. Levine, V. R. Brown, & T. Shirey (Editors), Oscillations in Neural Systems, Lawrence Erlbaum Associates, 2000.
- R. W. Parks, D. S. Levine, & D. L. Long (Editors), Fundamentals of Neural Network Modeling: Neuropsychology and Cognitive Neuroscience, MIT Press, 1998.
- D. S. Levine, & W. R. Elsberry (Editors), Optimality in Biological and Artificial Networks? Lawrence Erlbaum Associates, 1997
- D. S. Levine, & M. Aparicio, IV (Editors), Neural Networks for Knowledge Representation and Inference, Lawrence Erlbaum Associates, 1994
- D. S. Levine, & S. J. Leven (Editors), Motivation, Emotion, and Goal Direction in Neural Networks, Lawrence Erlbaum Associates, 1992
- Classification of Infrasound Events: A Neural Network approach
Prof. Fredric M. Ham, Ph.D.
Harris Professor of Electrical Engineering
Florida Institute of Technology
Abstract
Infrasound is a low-frequency acoustic phenomenon that occurs in nature, and can also result from man-made events, typically in the frequency range 0.01 Hz to 10 Hz. An integral part of the Comprehensive Nuclear-Test-Ban Treaty (CTBT) International Monitoring System is an infrasound-monitoring network. This network has the capability to detect and verify infrasonic signals-of-interest (SOI), e.g., nuclear explosions, from other unwanted infrasound noise sources, i.e., volcano eruptions, bolides, mountain associated waves, and microbaroms, to name a few. Results are presented for a bank of Radial Basis Function (RBF) neural networks, to discriminate between different infrasonic events. Each module in the bank of RBF networks is responsible for classifying one of several events, and thus, is trained to identify only this particular event. However, each module is also trained to not classify all other events. Output thresholds of each module are set according to specific Receiver Operating Characteristic (ROC) curves. Preprocessing of the infrasound signals is carried out by extracting cepstral coefficients and computing the associated derivatives that form the feature vectors used to train and test the RBF networks.
About the Plenary Speaker
FREDRIC M. HAM, PH.D., is Harris Professor of Electrical Engineering and Interim Dean of the College of Engineering. He worked 10 years in industry before coming to Florida Tech in 1988. He was with the Harris Corporation from 1980 to 1988 and the Shell Oil Company from 1976 to 1977. He is currently the Secretary of the INNS Board of Governors. Dr. Ham is a Senior Member of IEEE and has been an Associate Editor for the IEEE Transactions on Neural Networks since 2001. He has published over 100 technical papers and reports, mostly in the area of neural networks, but also in the areas of signal processing, controls and biomedical engineering (specifically, biosenors). He also holds 3 U.S. patents and is author of the book Principles of Neurocomputing for Science and Engineering, McGraw-Hill, 2001. He has served on many program committees for several conferences including, IJCNN, WCCI and SPIE. He is currently the Program Co-Chair for the IJCNN 2005 in Montreal, Quebec, Canada. Dr. Ham's current research interests include: applications of neural networks, recently for classifying infrasonic events, and development of non-invasive methods for monitoring blood glucose concentrations.
- Image Processing and Image Mining for Biotechnological Inspection Processes
Prof. Petra Perner
Computer Vision and Applied Computer Sciences
Leipzig Germany
Kornerstr. 10, 04107 Leipzig
ibaiperner@aol.com, www.ibai-institut.de, www.ibai-research.de
Abstract
There are many biotechnological applications where 3-dimensional objects are represented as 2-d objects in a digital image. The dynamic and variable nature of the microorganism thus creates a formidable challenge to the design of a robust 2-d image inspection system with the ideal characteristics of high analysis accuracy but wide generalization ability. We have developed a novel case-based object recognition method for this kind of problems. The method is able to recognize objects and learn incrementally cases for the recognition process. Such a procedure requires capturing new cases for the further recognition process in order to be able to handle the variability of the natural objects. Image mining is applied to the newly captured cases in order to keep the case base as small as possible. As a result we obtain groups of similar cases for which we are prototypically calculating cases that are stored into the case base. These learnt cases are applied for case-based object recognition. For the object recognition procedure we have developed a novel similarity measure that can determine similarity between the cases in the case base and the objects in the image. The similarity measure is flexible enough in order to get adjusted for different recognition purposes. We describe the theory behind it and how it works on our problem of fungi spore recognition. The developed case-based object recognition method is flexible and robust enough to be used for different recognition tasks in biotechnology.
About the Plenary Speaker
Petra Perner is the director of the Institute of Computer Vision and Applied Computer Sciences in Leipzig Germany. She received her Diploma degree in electrical engineering in 1981 and her PhD degree in computer science in 1985. From 1985 to 1991 she worked at the Technical University of Leipzig where she was the head of the research group "Opto-electronic Inspection and Diagnosis Systems". In 1991 she was a visiting scientist at the Fraunhofer-Institute of Non-Destructive Testing in Saarbrucken and in 1992 she worked as a visiting scientist at IBM T.J. Watson Research Center in Yorktown Heights, NY, USA. She was the chair of the Technical Committee 17 Machine Learning and Data Mining of the International Association of Pattern Recognition (IAPR). She has been the principal investigator of various national and international research projects. She received several research awards for her research work and has been recently awarded with 3 business awards for her work on bringing intelligent image interpretation methods into business. Her research interest is image analysis and interpretation, machine learning, data mining and case-based reasoning. Recently, she is working on various medical and biomedical applications and e-commerce applications.
- Novel Multifusion Imaging Technologies: From Cancer Detection to Semiconductor Wafer Inspection and Nanophotonics
Prof. George Constantine Giakos
Director Imaging Technologies, Photonics, and Applied Nanosciences Laboratories
Electrical and Computer Engineering Biomedical Engineering
Department of Electrical and Computer Engineering
The University of Akron
Auburn Science and Engineering Center
ABSTRACT
The rapid progress in imaging technologies during the last decades has stimulated many developments and applications in medicine and biology, defense, industry, aerospace, homeland security, remote sensing, meteorology, oceanography, and environment.
Imaging technologies from Space research have been identified, successfully applied, or explored towards the development of novel high-resolution, multi-sensor medical imaging technologies, defense, and homeland security applications. Similarly, experimental research findings for defense applications, have been applied towards the development of multifusion, optical sensing imaging systems and techniques, for efficient tumor detection, physiological imaging, and nanoimaging.
A major research breakthrough is the invention of the multifusion, multispectral, dual-energy, polarimetric imaging sensing platform. One application of the above technology is the detection and imaging of specific molecular signatures in vivo providing physiological and metabolic information at the molecular level. Other applications are in the areas of Space exploration, for robotic rovers vision, and well as nanoimaging,homeland security and defense for enhanced target recognition, identification, and surveillance, and high resolution imaging techniques for inspection, characterization, classification, and monitoring of wafers and masks for the microelectronic industry.
Another research breakthrough is the invention of multimedia-multidensity-multiatomic dual-energy imaging sensors. Dual-energy imaging involves the use of two x-ray images, one produced from a high energy and another from a low energy polychromatic spectrum. A weighted difference of these two images produces a digital image which eliminates interfering background structures. Simplifying the background structure leads to an increase in the detectability or "conspicuity" of the target structure. Utilizing two separate detection media allows the detector to be optimized for a wide range of applications. Applications of this technology are being explored for enhanced flat panel digital radiography, where cancellation of the displayed contrast of any two materials, such as bone and soft tissue, allow low-contrast lesions to be obtained through increased conspicuity in bone-canceled images, airport or building security, detection of explosive materials and airport luggage inspection and screening.
Overall, technology transfer, utilization, and exchange of the various imaging technologies will create advanced, integrated solutions for potential development in different areas such as medical diagnosis, defense, industry, and Space exploration. New imaging technologies will emerge and are expected to play an ever-expanding role in the technological arena.
About the Plenary Speaker
Dr. Giakos received his Degrees from Universita di Torino (Laurea), University of Edindurgh (Grad. Diploma), Ohio University (MS) and Marquette University (Ph.D.), specializing in Applied Physics, Nuclear Instrumentation, Nuclear Space Physics, Electrical Engineering, respectively. He did PostDoctoral Training at the University of Tennessee, specializing in the design of Medical Imaging Technologies. He is an Associate Professor in the Department of Electrical and Computer Engineering, with a joint appointment in the Department of Biomedical Engineering, at the University of Akron, OH. In addition, he is the Director of the Imaging Technologies, Photonics, and Applied Nanosciences Laboratory, a dedicated carrier of innovative ideas and inventions, with emphasis on the molecular imaging of cancer, defense, homeland security, and semiconductor wafer chip inspection and monitoring. While a faculty of the University of Akron, he developed and taught 20 distinct graduate and undergraduate courses, and secured sponsorship for several research projects. Currently, he is engaged on several research projects related to the exploration, design, and development of efficient imaging technologies with emphasis on molecular imaging and cancer detection, Space Exploration, inspection, characterization, classification, and monitoring of wafers and masks for the microelectronic industry, imaging solutions for defense and Homeland security. He is leading several interdisciplinary, multi-institutional efforts for funded research. Dr. Giakos has successfully competed for grant proposals and attracted funding from several NASA Agencies, Air force, Cleveland Clinic, Ohio Aerospace Institute, Lockheed Martin, Ohio Board of Regents, e-V Semiconductor Products, and several other agencies and companies. He have been instrumental to contribute towards the development of the Ohio BioMEMS Consortium on Medical Therapeutics Microdevices, and The Ohio Cellular and Molecular Imaging Consortium. Dr. Giakos is an Associate Editor of the IEEE Instrumentation and Measurement Transactions and the Chairman of the IEEE Technical Committee on Imaging Systems and Techniques. Dr. Giakos research has fostered several breakthrough inventions which have been rewarded with seven (10) US Patent Awards, several other pending patents, and more than 120 peer-review articles and journal publications.
- Inverse scattering procedures for active imaging systems at radio frequencies and microwaves
Prof. Matteo Pastorino
Department of Biophysical and Electronic Engineering
University of Genoa, Italy
ABSTRACT
Systems for electromagnetic imaging at radio frequencies and microwaves are of great interest in several applications, such as industrial and civil non-destructive evaluation and testing (NDE/NDT), subsurface inspection, and medical imaging.
At those frequencies, the ratio between the linear dimensions of the object and the wavelength is such that the relationships between the measured data (values of the scattered electric field) and the unknown properties of the object (e.g., shape, dimensions, distributions of dielectric parameters) are governed by scattering laws. This is the key point making the development of efficient imaging systems working at these frequencies, at the same time, difficult and appealing. Difficulties are mainly related to the fact that simplified hypotheses (e.g., ray propagation) cannot be applied and the information about the object under test is contained in a complicated way in the measured data. On the contrary, short-range electromagnetic imaging can provide, in principle, information on the electric properties of the objects under test (e.g., the distributions of the dielectric permittivity, electric conductivity, magnetic permeability, etc.), which cannot be directly deduced by using other diagnostic techniques. Moreover, the equipments required to develop electromagnetic imaging systems are far less expensive comparing with other currently used inspection techniques (e.g., CT, NMR, PET). These potential advantages of electromagnetic imaging could be very interesting for the new generation of imaging apparatuses. It is generally recognized that noinvasive diagnostic techniques are expected to play an ever-expanding role in different domains of science and technology. In this context, besides consolidate diagnostic approaches of high intrinsic value, microwave techniques can also be of interest due to their potential capabilities of monitoring and imaging highly contrasted objects continuously and at low dose.
In this context, the development of accurate and efficient inversion and reconstruction procedures represents a critical point for microwave imaging systems, due to the well-known nonlinearity and ill-posedness nature of the inverse scattering equations. In this paper, the above-mentioned key issues are discussed and some of the recently proposed inversion methods for two-dimensional tomographic imaging are briefly reviewed. In particular, the paper is focused on both deterministic and stochastic inversion procedures, which have been proposed to solve the inverse scattering problem with reference to both its "exact" formulation and by using linear and nonlinear approximations (e.g., the first- and second-order Born approximations).
About the Plenary Speaker
Matteo Pastorino received the "laurea degree" (Master Degree) in Electronic Engineering from the University of Genoa in 1987 and the Ph.D. degree in Electronic Engineering and Computer Science from the same university in 1992. At present, he is the vice director of the Department of Biophysical and Electronic Engineering, University of Genoa, Italy, where he is a professor of Electromagnetic Fields, Antennas and Remote Sensing. He is also the director of the Applied Electromagnetics Group. His main research interests are in the field of imaging systems, direct and inverse scattering problems, industrial and medical applications, smart antennas, and analytical and numerical methods in electromagnetism. He is co-author of more than 250 papers in international journals and proceedings of conferences. Prof. Pastorino is a Senior member of IEEE, member of the IEEE I&M Tech. Committee on Imaging Systems, co-chair of the IEEE Workshop on Imaging Systems and Techniques. He is also a member of the review and editorial boards of some international journals in the field of microwaves and antennas and a member of the Italian Electromagnetics Association (SIEM).
- Transcoding MPEG-2 compressed video into H.264/AVC
Dr. Daniele Bagni
Multimedia Coding SW Expert
Advanced System Technology labs
STMicroelectronics, Agrate Brianza, Italy
Abstract
The emerging H.264/AVC standard (also known MPEG-4 part 10) achieves much higher compression at the same image quality than MPEG-2, MPEG-4 and H.263 and has a wider application spectrum in comparison with previous standards, ranging from broadcast HDTV transmission, to HD DVD, mobile phone and video streaming. For all these new applications, H.264/AVC will probably become the choice in the incoming years. Nevertheless, we have to take into account the huge amount of previously created, broadcasted or stored MPEG-2 material, during the last ten years of MPEG-2 “dominance” in digital video transmission at mid-high bitrates.
To maintain compatibility we developed a “transcoding” system, that is, a conversion process of transforming a compressed MPEG-2 video bitstream into an H.264/AVC bitstream, without necessarily doing a full MPEG-2 decoding step followed by a full H.264/AVC re-encoding one. The scheme of our transcoding is composed of two main stages: a complete MPEG-2 decoder and a “light” H.264 encoder stage. We call it DBS, or Dynamic Bitstream Shaper, since it achieves syntax translation as well as bitrate compression.
Our DBS works as a “fast encoder”, where all the most important H.264 coding decisions are inferred from the MPEG-2 input bitstream: during the MPEG-2 decoding phase we store and pass decoded parameters quantities like motion vectors, quantization factors and coding modes to the H.264 stage, thus avoiding a complete motion estimation process.
In terms of performance, our experiments show a compression gain between 25% and 40% respect to the MPEG-2 input bitrate. We have also measured that our quality (PSNR) is at least equal or better that the full decode/re-encode (“trivial”) solution. The implementation complexity of the DBS is anyway at least one order of magnitude lower than the full chain, due to re-using the information arriving from the input bitstream. Such a re-use allows us to skip the Motion Estimation process, which is by far the most CPU demanding part of the H.264/AVC standard.
About the Plenary Speaker
Daniele Bagni was born in Ferrara, Italy in 1963. He graduated in Quantum Electronics at the Milan Polytechnic University, Italy, in 1990, after one year in STMicroelectronics as designer of analog circuits. He then joined Philips Research laboratories in 1991, working on real-time hardware prototyping of advanced motion compensated scan-rate conversions. He also operated on H.263 low bit-rate video coding based on a VLIW media-processor. He joined STMicroelectronics’ Advanced System Technology labs at the end of 1997, where he is currently working on multimedia algorithms and related system-architecture making use of VLIW cores. He holds patents granted in Europe and USA in digital video processing. From 1999 he is also external professor at the State Milan University, Information Science department, where he is teaching “Multimedia Information Coding”.
- Frequency Domain Adaptive Filtering in Signal Processing and Communications
Prof. Kosta Berberidis
Computer Engineering & Informatics, School of Engineering,
University of Patras, Greece
Abstract
In recent years there is an increasing interest in adaptive signal processing algorithms which are implementable in the frequency domain. Due to their computational efficiency and their good convergence properties, Frequency Domain Adaptive Filtering (FDAF) algorithms turn out to be among the most efficient solutions in several practical situations.
In particular, FDAF algorithms are very useful in real-time applications involving long adaptive filters.
Most of the existing FDAF algorithms are of the gradient type, that is, their time-domain counterparts are based on Least Mean Square type algorithms. However, there have been some recent efforts with promising results towards deriving frequency domain implementations of Quasi-Newton algorithms as well.
The aim of this paper is to present a review of Frequency Domain Adaptive Filtering with focus on the basic ideas and tools that lead to efficient implementations. Also, modern real-time applications in signal processing and communications will be discussed, such as, Acoustic Echo Cancellation, Channel Estimation, and Channel Equalization in Single-Carrier Communication systems.
About the Plenary Speaker
Kostas Berberidis was born in Serres, Greece, in 1962. He received the Diploma degree in electrical engineering from the Democritus University of Thrace, Greece, in 1985, and the Ph.D. degree in signal processing and communications from the University of Patras, Greece, in 1990.
From 1986 to 1990, he was a Research Assistant at the Computer Technology Institute (C.T.I.), Patras, Greece and a Teaching Assistant at the Computer Engineering and Informatics Department, University of Patras.
During 1991, he served in the Greek Army working at the Speech Processing Lab of the National Defense Research Center. From 1992 to 1994 and from 1996 to 1997, he was a researcher at the R&D department of C.T.I. During academic year 1994/95 he was a Postdoctoral Fellow at C.C.E.T.T. (Centre Commun d'Etudes de Telediffusion et Telecommunications), Rennes, France.
Since December 1997, he has been with the Department of Computer Engineering & Informatics, School of Engineering, University of Patras, where he is currently an Associate Professor and Head of
the Signal Processing and Communications Lab.
His research interests include fast algorithms for adaptive filtering, direction-of-arrival estimation and tracking and wireless communications.
Dr. Berberidis was a member of Scientific and Organizing Committees of several International Conferences and he is currently serving as Associate Editor of the EURASIP Journal on Applied Signal Processing. He is also a member of the Technical Chamber of Greece.
- Knowledge-Based Web Systems for Product Representation and Recommendation
Prof. Bhanu Prasad
Department of Computer and Information Sciences
Florida A&M University
Abstract
Knowledge-based approaches are extensively used in the development of product representation and recommendation systems. This talk focuses on various knowledge-based approaches such as case-based reasoning, automated-collaborative filtering, agents, etc., that are used in the development of successful product representation and recommendation systems. The talk also includes some research conducted by the author and his team at Florida A&M University, USA. Product representation and recommendation is very important in business-to-customer (B2C) e-commerce and other web related activities.
About the Plenary Speaker
Dr. Bhanu Prasad is an Assistant Professor in the Department of Computer and Information Sciences at Florida A&M University (website: http://www.cis.famu.edu/profile/profile-prasad.html). He received Masters and Ph.D. degrees, both in computer science, from Andhra University (India) and Indian Institute of Technology Madras (India) respectively. In the past, he worked with three multi-national software companies. Dr. Prasad’s research is primarily focused on the theory and applications of knowledge-based and agent-based systems. He is/was as a chair, co-chair, program committee member, keynote speaker, and guest editor for several international conferences and journals.
- Superfast Filters
Prof. Dr. Demetrios G. Lainiotis
Abstract
The superfast Lainiotis filter is presented. This multi-model filter is an exact realization of the optimal use estimator for linear gaussian models. The Lainiotis filter has its most superfast nature in its distributed processing realization. This property constitutes a significant advantage over competitive filters, e.g. the Kalman filter. For example, its competitive advantage over the Kalman filter exceeds 10 orders of magnitude for time-varying, as well as, for time-invariant problems.
Moreover, in addition to its optimality and superfast nature, the Lainiotis filter possesses many additional significant advantages one of which is its natural parallel processing/distributed processing structure, and its ease of implementation in software or hardware.
The Lainiotis filter has been applied to numerous important applications in signal processing, communications, optimal control, etc.
About the Plenary Speaker
Dr. Lainiotis has served as Professor of Electrical Engineering, and Associate Director of the Electronics Research Center at the University of Texas at Austin, and as Professor and Chairman of the Electrical Engineering Department of the State University of New York at Buffalo. He has also held the Chair of Pattern Recognition at the University of Patras, Greece 1981-1990 from which he resigned in order to accept the Harris Corporation Endowed Chair Professorship at Florida Institute of Technology. Since 1994 he has been retired.
Professor Lainiotis is the developer and innovator of the Multi-Model theory and technology that is of scientific significance and extensive practical applicability to engineering areas of economic value and importance, such as telecommunications, biomedical signal processing, robotics, airplane/ship automation, oil exploration, structural engineering, etc. The Multi-Model methodology constitutes a powerful unifying framework for signal processing (filtering, forecasting), decision (detection, pattern recognition), and control. It unifies and incorporates the statistical and neural network approaches, as well as the fuzzy set methodology. It also serves as the fundamental framework for adaptive/learning systems, and for intelligent or self-organizing systems. Based on the Multi-Model methodology, very efficient and superfast algorithms have been obtained for signal processing filtering (nonlinear, linear adaptive), and for pattern recognition, and control.
Dr. Lainiotis has served as Associate Editor of the IEEE Transaction on Signal Processing, IEEE transactions on Automatic Control, the IEEE Transactions on System, Man, and Cybernetic, the International Journal of Information Sciences, the Journal of Pattern Recognition, the Journal of Neural, Parallel and Scientific Computations, and the Journal of Mathematical Problems in Engineering. He also served as Guest Editor of the Proceedings of the IEEE, Special Issue on Adaptive Systems, August 1976, of the Journal of Information Sciences, Special Issue on Estimation, Fall 1974, and of the IEEE Transactions of Geoscience Electronics, Special Issue on Geophysical Signal Processing, Jan. 1977.
|
NEWS
|