Michalis Zervakis

Director

Digital Image and Signal Processing Laboratory (DISPLAY)

Department of Electronics and Computer Engineering

Technical University of Crete (TUC)

Chania 73100

Crete

GREECE

Telephone : +30-821037206
Telefax :         +30-8210375422
michalis@danai.systems.tuc.gr

Main activities:

The Technical University of Crete is one of the foremost Engineering schools in Greece , pursuing education and research in modern aspects of Technology. The Department of Electronics and Computer Engineering has 18 academic staff, more than 500 undergraduate and 80 graduate students. The DISPLAY group is R&D group within the Department of Electronic and Computer Engineering and has extensive experience in a variety of signal & image processing issues. Representative work of the groups appears in the following areas:

·     Image Processing and Analysis

·     Nonlinear Identification using Neural Networks and Fuzzy Systems

·     Neuro-fuzzy Modelling

·     3-D Modelling

·     Biomedical Image Processing

·     Time-Series Processing

·     Fault Monitoring

·     Product Inspection and Quality Control

·     Adaptive and Predictive Control

·     Multimedia Information Systems

·     Internet Information Systems

·     Spatial Access methods, Access Methods for images and Video

·     Virtual Reality Software Algorithms

·     Video Processing and Compression

·     Renewable Energy Sources

DISPLAY has a large experience on several research and development projects with funding from the European Commission and the Greek goverment.  Some indicative activities of our group in such projects are outlined in the following

• Title: "High Performance Industrial Vision", RTD Project. Source: BRITE-EURAM III European Research Programme. Period: 1998 - 2001.

• Title: “Advanced Tomographic Sensors for Industrial Multiphase Imaging”. Source: BRITE-EURAM III European Research Programme. Period: 1998-2001

• Title: “Dissemination of Advanced Control Technologies for Isolated Power Systems with Increased Renewable Energy Sources Penetration, Source: THERMIE B, Period: 1998 - 2001

• Title: "MEDEA - Quality Control for House-hold Appliances by On-line Evaluation of Mechanical Defects, Source: SMT EE Programme, Period: 1995-1998.

• Title: "Wastepaper Classification and Processing". Source: Superior Recycled Fiber Industries (SRFI) of Minnesota , Period: 06/01/93 - 09/30/93

• Title: "Characterisation of Contaminants in Waste Paper Bales Based on Physical Charecteristics of Contaminants, Source: Superior Recycled Fiber Industries (SRFI) of Minnesota , Period: 06/01/94 - 09/30/94

• Title: "Development of a Weather Information System", Source: Minnesota , Department of Transportation, Period: 07/0l/94 - 09/30/94 .

• Title: "Development of an Automated Robotic System with applications in biomedical Operations", Source: Distinct of Crete, Development fund, Period: 0l/0l/96 -12/31/97

• Title: "Development and Implementation of an advanced control system for the optimal operation and management of medium size power systems with a large penetration from renewable power sources", Source: JOULE II EE Programme, Period: 09/01/92 - 01/30/93

The DISPLAY group of the Technical University of Crete can mainly provide its expertise in the A3, S2, T1, T3, T4, D1 and D2 JPAs of the PATIMON NoE. TUCs proposal with regards to its contribution to the above JPAs is described below:

A3: Communication CAMS-CAMS for knowledge transfer

Support of a distributed data management and retrieval system implemented over the internet, supporting high level of data integration (any information source is accessible from anywhere) as well as intelligent data processing and retrieval functions such as content-based access of signals and images. The communication takes advantage of the existing infrastructure of the internet but also supports efficient transcoding and retrieval. Existing standards for looseless and lossy compression are adopted (e.g., MPEG) and also standards for metadata information representation and communications (e.g., XML).

S2: Knowledge Management and DataBase System

It is important to extend the capabilities of the PatiMan DataBases system by developing database systems supporting the automated management and retrieval of medical signals by content. For example, given a patient exam (e.g., CT, X-ray etc) this will allow the physician to retrieve similar examination from the medical archive. This functionality can become an  important component of the overall Knowledge management and database system and support administrative, clinical, teaching and research activities. Retrieval of similar examinations would not only reveal cases of similar examinations with diagnosis but also cases of similar examinations with different diagnoses. Databases design, information retrieval, data representation and inference are also issues to be examined.

In addition, called “information layer” of PatiMon covers the attributes of information as used by technology to achieve specific objectives and interacts with the Data and Technology layers, as technology uses and generates new information along the process of implementing application-specific functionalities. It defines an information hierarchy of the information which is either initially available to an application (e.g., raw data such as images, signals, dates etc) or is computed by various information processing steps. It mainly consist of three levels (a) low level containing mainly raw data (signals, images, dates, names etc), (b) intermediate level defined mainly on computed features and (c) high level consisting mainly of metadata rich in semantic content, directly associated with the goals of each application.

T1: Microsystems and Wireless Sensors

Several researchers of the Display lab have extensive expertise on electronic and microsystem devices, smartcards, biosensors, devises for persons with special needs, etc. This expertise can be utilized in all development phases, from design to implementation and testing.

T3: Transmission and interfaces, security and data transmission

The communications takes advantage of the existing infrastructure of the internet but also supports efficient transcoding and retrieval. Existing standards for looseless and lossy compression are adopted (e.g., MPEG) and also standards for metadata information representation and communications (e.g., XML). Design of intelligent user interfaces over the network is also an important issue here.

T4: Signal Processing, classification, recognition and Evaluation

Design and analysis of intelligent (e.g., adaptive) algorithms for recognizing and classifying multimodal signals and revealing associations between them. Data fusion, interpretation and data mining are issues can be explored by TUC. An important aspect of the knowledge management system has to do with the mining of information, that is, identification of conceptual similarities, clustering of common behavior, etc.).

Key personnel

The Display group will participate in PatiMon with five (5) researchers and five (6) doctoral students. Below are the names and expertise of the researchers while the names of the doctoral students will be defined in due time.

Researchers

Michael Zervakis holds a Ph.D degree from the University of Toronto , Department of Electrical Engineering, since 1990. He joined the Technical University of Crete on January 1995, where he is currently full professor at the department of Electronic and Computer Engineering. Prof. Zervakis is the director of the Digital Image and Signal Processing Laboratory (DISPLAY) and is involved in research on modern aspects of signal processing. He has been involved in more than 20 international projects and has published more than 60 papers in related areas of image/signal processing.

Euripides Petrakis received the PhD degree in Computer Science from the University of Crete, Greece in 1993.  He is currently an associate professor at the dept. of Electronic and Computer Engineering of the TUC. His research interests include image and video databases, computer vision, image processing, databases, access methods for image data and video and their applications to the retrieval of multimedia documents by pictorial content. His is currently involved in research for multimedia data management and mining on the Internet.

George Stavrakakis received the PhD Degree in Systems Automation and Control from the PAUL SABATIER, Toulouse III University of Toulouse in 1983. He is currently a professor at the dept. of Electronic and Computer Engineering of the TUC. His research interests include application of the theory of evidence and estimation in industrial and aerospace environments, production systems automation, decision analysis, real time fault monitoring of industrial processes, fuzzy logic etc.

Costas Kalaitzakis received his PhD in Electronics from the Democritus University of Thrace in 1983. He is an Associate Professor and Chairman of the Department of Electronics and Computer Engineering of the Technical University of Crete, Greece. His research interests include electronics, sensor and measurement systems, smart card applications, bioengineering and local operating networks.

Spyreta Golemati received the PhD Degree in Biophysics, Medical Physics and Bioengineerin from the Imperial College of Science, Technology and Medicine, University of London , UK in 2000. She is currently a research scientist at the Dept. of Electrionic and Computer Engineering. Her research interests are in the area of medical imaging and biomedicine.

Doctoral Students

Michael Kounelakis

Ioannis Dimou

Vaggelis Sakalis

Michael Blazantonakis

Doctoral student

Doctoral student

Recent Publications  

[1]  G.A. Rovithakis, M. Maniadakis, and M. Zervakis, “Artificial Neural Networks for  Feature Extraction and Classification of Vascular Tissue Fluorescence Spectrums”, ICASSP 2000, June 5-9, 2000, Instabul, Turkey.

[2]  A. Matamis, G.A. Rovithakis, and M. Zervakis, “Controling QoS at the Application Level for Multimedia Applications using Artificial Neural Networks: Experimental Results’, EUSIPCO 2000, Tampere Finland, Sept. 4 – 8, 2000.

[3]  M. Zervakis, V. Sundararajan and K.K. Parhi , “Robust Image Denoising in the Wavelet Domain”, Journal of Image and Vision Computing, Elsevier, 2001.

[4]  S. Mertikas and M. Zervakis, “Exemplifying the Theory of Evidence in Remote Sensing Image Classification, International Journal of Remote Sensing, 2001.

[5]  S. Goumas, M. Zervakis, A. Pouliezos and G. Stavrakakis, “Intelligent On-Line Quality Control of Washing Machines using Discrete Wavelet Analysis Features and Likeligood Classification”, Engineering Applications of Artificial Intelligence, Elsevier, 2001.

[6]   Elias N. Malamas, Euripides G.M. Petrakis, Michalis Zervakis, Laurant Pettit and Jean-Didier Legat, "Industrial Vision: Systems, Tools and Techniques", Image and Vision Computing (accepted, to appear).

[7] Euripides G.M. Petrakis, Aristeidis Diplaros and Evangelos Milios: "Matching and Retrieval of Distorted and Occluded Shapes using Dynamic Programming", IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 24, No. 11, November 2002, pp. 1-16.

[8] Euripides G.M. Petrakis: "Fast Retrieval by Spatial Structure in Image DataBases", Journal of Visual Languages and Computing, Vol. 13, No. 5, October 2002, pp. 545-569.

[9] Euripides G.M. Petrakis: "Design and Evaluation of Spatial Similarity Approaches for Image Retrieval", Image and Vision Computing, January 2002, Number 1, Volume 20, pp. 59-76.

[10] Euripides G.M. Petrakis: "Content-Based Retrieval of Medical Images", International Journal of Computer Research, Volume 11, Number 2, 2002, pp. 171-182.

[11] Euripides G.M. Petrakis and Kostas Tzeras: "Similarity Searching in the CORDIS Text Database", Software Practice and Experience, Nov. 2000, No.  13, Vol. 30, pp. 1447-1464.