Euripides G.Μ. Petrakis
Position: Professor, Lab. Director
Office: 145.A29
Phone: +30-28210-37229
Fax: +30-28210-37542
Email: petrakis intelligence.tuc.gr
Personal Page: http://www.intelligence.tuc.gr/~petrakis/
Office: 145.A29
Phone: +30-28210-37229
Fax: +30-28210-37542
Email: petrakis intelligence.tuc.gr
Personal Page: http://www.intelligence.tuc.gr/~petrakis/
Short CV
Prof. Euripides Petrakis received Bachelor in Physics from the National University of Athens in 1986. He received a Ph.D in Computer Science from University of Crete, Dept. of Computer Science, since 1993. Between 1996 and 1998 he was a visiting professor at the Dept. Computer Science of York University, Toronto, Canada and a researcher at GMD/IPSI Institute, Darmstadt, Germany. He was awarded with an ERCIM fellowship in 1998. He is serving as professor at the Computer Science division, of Electrical and Computer Engineering Dept. of the Technical University of Crete (TUC), and since 2006, as director of the Intelligent Systems Laboratory. Prof. Petrakis is involved in research on computer vision and image databases (at the early stages of his career), on modern aspects of information systems (Web information systems, semantic Web) and recently, cloud computing and IoT. Prof. Petrakis has authored or co-authored over 120 papers with students he has supervised (as main or sole supervisor) in high quality journals and conferences (acceptance rate < 30%) and has over 4000 references to his published work (by Google Scholar). He is involved (as principal investigator or coordinator) in several research projects funded by the Greek Government and the EU and attracted over 2.5Meuros funding for TUC.Education
- 1993: Ph.D. in Computer Science from the Department of Computer Science, University of Crete.
Title: "Image Representation, Indexing and Retrieval based on Spatial Relationships and Properties of Objects" (supervisor: Stelios C. Orphanoudakis) - 1985: B.Sc. in Physics from the Department of Physics, National University of Athens
Research Interests
- Multimedia, Medical and Web Information Systems
Information systems technology with emphasis on the management (processing, information extraction, dissemination and retrieval) of complex information. As such, a database is not only considered from a standalone perspective; its strength lies in its potential association with other technological approaches e.g., web, content-based approaches.
[More information about Multimedia, Medical and Web Information Systems]
- Semantic Web
Our work aims at enhancing the capabilities of state-of-the-art information systems and their support for information analysis and reasoning by exploiting the time dimension in the information possessed. Knowledge is being interpreted through Time Determined Ontology (TDO) models encoding concepts and relationships acquired by experts and by data analysis. It is developed based on state-of-the-art natural language processing, data analysis (including signal and image analysis) and semantic web technology (e.g., OWL, OWL reasoners, query languages) thus exploiting leading edge technology in knowledge management.
[More information about Semantic Web]
- Natural Language Processing (NLP) and Information Extraction
Information extraction from text sources, mining concepts and relationships for ontology construction in information systems design and development. This work permits informal textual descriptions of multimedia content to be effectively used in retrieval, and obviates the need for generating structured metadata. Informal descriptions require significantly less human labor than structured descriptions. This work permits enables also knowledge interpretation through ontologies and ontology-based and web-search engines.
[More information about Natural Language Processing (NLP) and Information Extraction]
- Access Methods for Images and Video
Information retrieval, Information systems, Multimedia information systems, Web Information systems, Image/Spatial Databases, Image content analysis, Image indexing, Retrieval by image content.
[More information about Access Methods for Images and Video]
- Applications of Computer Vision
Image / video processing and compression, Image / Video content analysis and description (e.g., motion analysis, shape analysis, object tracking), Pattern recognition. Design and analysis of Inspection Systems; Panoramic image generation; Real-time implementation of integrated imaging systems on DSP boards.
[More information about Applications of Computer Vision]
Teaching
- AIS 413: Multimedia Data Management
Description
Processing, archiving, and searching multimedia information including documents, one-dimensional signals, still and moving images (video) in information systems and the Internet. Classic models of information retrieval (binary, relational, probabilistic), information clustering and clustering algorithms (partitional, hierarchical, hybrid algorithms), clustering applications grouping in document collections. Visualization of one-dimensional signals and images in multimedia systems. Feature extraction (color, texture, shape, and spatial relationships) from images. Retrieval methods for one-dimensional signals and images. Indexing techniques in information systems for documents and multimedia information (inverted files, k-d trees, grid files, R-trees). Design of information systems on the Internet, management and analysis of information on the Internet (PageRank and HITS methods). Basic processing techniques and analysis of still and moving images (video) in information systems. Compression techniques, JPEG, MPEG-1, 2, 4, 7 standards.
Courses Portal Link: Multimedia Data Management
- AIS 414: Machine Vision
Description
Basic principles and methodology of machine vision with emphasis on algorithms and applications of machine vision. Image formation, mathematical, geometric, colour, frequentist, discrete models. Basic image processing techniques (filtering, enhancement, normalization). Edge detection, first and second derivative operators. Image segmentation, methods for segmenting or enhancing regions and edges, thresholding techniques. Advanced segmentation techniques (merging and splitting regions and edges, relaxed ordering, Hough technique). Techniques for processing binary images, distance transforms, morphological operators, and region labeling. Analysis, representation, and recognition of images. Representation of edges and regions, representation and recognition of shapes, representation and recognition of structural content. Texture analysis and recognition, structural and statistical methods. Dynamic vision, estimation of motion, optical flow, and trajectory.
Courses Portal Link: Machine Vision
- COMP 211: Data Structures and Algorithms
Description
Abstract Data Types, implementation in Java, algorithm complexity, performance analysis of algorithms. Sorting in main and external memory, sorting algorithms: bubble sort, exchange sort, insertion sort, selection sort, quick sort, merge sort, k-way merge sort, radix sort. Stacks, queues, linked lists. Implementation of one-dimensional arrays and dynamic memory allocation. Trees, tree traversal, binary search trees, operations research in binary trees (search, insert, delete data). Implementation using arrays and dynamic memory allocation. Applications, Huffman codes. Graphs, graph traversal. Operations on graphs (search, insertion, deletion). Implementation of graphs and applications (minimum spanning tree, shortest path). Searching in main or external memory. Sequential search (binary search, interpolation search, self-adjusting search), Indexed sequential search, ISAM. Performance analysis of search. Hierarchical search trees, trees in main memory (binary search trees, AVL trees, optimal trees, splay trees), analysis of performance. Trees on the secondary memory (multi-way search trees, B-trees, B +-trees), VSAM. Tries, digital search trees, text tries, Patricia tries, Ziv-Lembel coding. Searching in text (KMP, BMH algorithms). Non-hierarchical search, hashing in the main memory, collision resolution, open addressing, separate chaining. Complexity of search. Hashing in external memory (dynamic hashing, extendible hashing, linear hashing). Performance analysis of search.
Courses Portal Link: Data Structures and Algorithms
- AIS 603: Multimedia Data Management
Description
Processing, archiving, and searching multimedia information including documents, one-dimensional signals, still and moving images (video) in information systems and the Internet. Classic models of information retrieval (binary, relational, probabilistic), information clustering and clustering algorithms (partitional, hierarchical, hybrid algorithms), clustering applications grouping in document collections. Visualization of one-dimensional signals and images in multimedia systems. Feature extraction (color, texture, shape, and spatial relationships) from images. Retrieval methods for onedimensional signals and images. Indexing techniques in information systems for documents and multimedia information (inverted files, k-d rees, grid files, R-trees). Design of information systems on the Internet, management and analysis of information on the Internet (PageRank and HITS methods). Basic processing techniques and analysis of still and moving images (video) in information systems. Compression techniques, JPEG, MPEG-1, 2, 4, 7 standards. Video segmentation into shots, shot aggregates.
Courses Portal Link: Multimedia Data Management
- COMP 607: Machine Vision
Description
Basic principles and methodology of machine vision with emphasis on algorithms and applications of machine vision. Image formation, mathematical, geometric, colour, frequentist, discrete models. Basic image processing techniques (filtering, enhancement, normalization). Edge detection, first and second derivative operators. Image segmentation, methods for segmenting or enhancing regions and edges, thresholding techniques. Advanced segmentation techniques (merging and splitting regions and edges, relaxed ordering, Hough technique). Techniques for processing binary images, distance transforms, morphological operators, and region labeling. Analysis, representation, and recognition of images. Representation of edges and regions, representation and recognition of shapes, representation and recognition of structural content. Texture analysis and recognition, structural and statistical methods. Dynamic vision, estimation of motion, optical flow, and trajectory.
Courses Portal Link: Machine Vision
Undergraduate Courses
Postgraduate Courses
Supervision
Publications
Funding
- FI-STAR: Future Internet Social and Technological Alignment Research
FI-STAR (Phase 2) is establishing early trials in the healthcare domain building on Future Internet technology. It will create a robust framework based on the ‘software to data’ paradigm and validate the FI-PPP core platform concept. FI-STAR will build a vertical community in order to create a sustainable ecosystem for all user groups in healthcare.
[More information about this project] - AI-CARE: Advanced Knowledge Management Systems with Inherence Computational Intelligence and Applications in Health Care
The AI-CARE project is expected to provide answers to relevant questions related to bipolar disorder and epilepsy issues, such as the individualization of diagnosis, treatment approaches and effectiveness of treatment, quality of life, transition hazard from major depressive episodes to manic, hypomanic, or mixed states, malignant types of bipolar disorder, psychiatric disturbances and suicide risk after epilepsy surgery and high risk patients with epilepsy.
[More information about this project] - MEDEVIPAS: Medical Devices Vigilance and Patient Safety
The aim is to timely extract the information on potentially hazardous MDs and make it available where appropriate. It is expected that the proposed project through the design, development and implementation of modern ICT tools for MDs vigilance, will create a prototype system, which will provide critical, on time and customised information to health care institutions, in addition to the existing traditional Systems and consequently will considerably improve patient safety.
[More information about this project] - eTRAWELSPA: e-TRAINING FOR WELLNESS IN SPA SERVICES
Promoting lifelong learning in the field of health is the result of continuous professional skills development through access to most recent information and a large knowledge data base. Latest international reports reveal that there is an increased therapeutic and social interest for health resorts’ spa treatments, which have led to the development of international balneary tourism. According to specialty EU statistics, the immediate results after treatment reflected significant improvements in subjective and objective symptomatology. Also an important echo was registered after using economical indicators of effectiveness, respectively on temporary incapacity to work, determined by the health conditions for which those spa treatments were being used. The aim is to identify, to diversify, to reorganize and to modernize the physical therapy strategies and tourist services using e-learning products from another LdV project to the level of treatment facilities within health resorts.
[More information about this project] - YPERTHEN: Computational Tools and Services for Diagnosis and Personalized Treatment of Cancer
Clinical Research Infrastructure and methodology for personalized tumor diagnosis and therapy
[More information about this project] - PoTrain: Positive Thinking Training
The Positive Thinking Training project aims to encourage the collection and the dissemination of positive information about the events that occur every day throughout the world. During this project young people will be encouraged to envision a better future for themselves and the world they live in.
[More information about this project] - RT3S: Real Time Simulation for Safer Vascular Stenting
RT3S will develop and validate a sophisticated patient-specific, probabilistic model of the fatigue-fracture of a stent, integrated in a computer-aided surgery planning application, implemented to run in real-time during the surgical planning, so as to provide advice of the risk of stent rupture while the surgeon is planning the operation. The real time software library, easy embeddable in any existing application, will make it possible to include the assessment of risk for stent fracture in all software solutions for computer-aided planning, training and intervention of peripheral vascular angioplasty procedures.
[More information about this project] - SOWL: Spatio-Temporal OWL for Reasining and Querying over the Semantic Web
Current technological tools for description of semantic knowledge are insufficient to support adequate automated reasoning in a dynamic (real-world) knowledge space. The objective of the project is to expand the current state of the art ontology languages (OWL, RDF-S, RDF) and their support for automated reasoning by adding a spatial dimension to enable real time context aware information analysis. This technology will provide ontology based (semantic) information systems to venture beyond a static world and add the concepts of space and change.
[More information about this project] - TOWL: Time-determined ontology based information system for real time stock
The objective of the project is to expand the current state of the art ontology languages (OWL, RDF-S, RDF) and their support for automated reasoning by adding the time dimension to enable real time context aware information analysis. This technology will provide ontology based (semantic) information systems to venture beyond a static world and add the concepts of time and change.
[More information about this project] - GCB: Getting Cretan Businesses On-Line and Doing E-Business
The role of this laboratory is to train consultants that will advise Cretan businesses about the benefits of getting on-line and doing e-business.
[More information about this project] - ICS-AMD: Intelligent Camera Systems supporting Automated Motion Detection
Integrated system of interconnected (through LAN) high definition cameras with motion detection support
[More information about this project] - MedSearch: Semantic Similarity for Knowledge Discovery in Medical Archives
Information Retrieval System for MedLine and Medical Information. MedSearch is a complete retrieval system for Medline, the premier bibliographic database of the U.S. National Library of Medicine (NLM). MedSearch implements the Semantic Similarity based Retrieval Model (SSRM), a novel information retrieval method for discovering similarities between documents containing semantically similar but not necessarily lexically similar terms.
[More information about this project] - OpTag: Improving Airport Security and Passenger Flow by Enhanced Passenger Monitoring
This project aims to harness emerging passenger tracking and identification technologies with the objectives of increasing the safety of air travel whilst maximising the utilisation of existing facilities.
[More information about this project] - BIOPATTERN: Computational Intelligence for Biopattern Analysis in Support of eHealthcare
The BIOPATTERN Network of Excellence is a groundbreaking project that integrates key elements of European research to enable Europe to become a world leader in eHealth. The Grand Vision is to develop a pan-European, coherent and intelligent analysis of a citizen's bio-profile; to make the analysis of this bio-profile remotely accessible to patients and clinicians; and to exploit bio-profile to combat major diseases such as cancer and brain diseases.
[More information about this project] - HERAKLETOS: Program for Basic Doctoral Research
Program for funding PhD candidates.
[More information about this project] - MULTI-MINE: Multimedia Data Management and Mining
Education and Training
[More information about this project] - HIPER: High performance industrial vision
Design and development of the software of the CMOS camera for industrial vision applications.
[More information about this project]