|
|
| Presentation |
|
Tomislav Buljubašić (Siemens, Zagreb, Croatia)
Establishing Innovative Culture in IT Organization |
| G1 |
T. Rolich (Tekstilno-tehnološki fakultet, Zagreb, Croatia), T. Rolich, D. Grundler (University of Zagreb, Faculty of Textile Technology, Zagreb, Croatia)
Determining Optimal Power, Location and Direction of Transmitters Using a Genetic Algorithm 
Electromagnetic field pollution caused by various human-made sources can pose a health hazard, especially in urban and dense populated areas. Many independent service providers install new electromagnetic sources (transmitters) usually without mutual coordination. In this paper an investigation is presented for adjustment of transmitters’ parameters (e.g. location, direction, power) to minimize a health hazard caused by the electromagnetic radiation while in the same time satisfy service availability (minimal reception electromagnetic field strength for each provider). In this research two transmitters with a different nonuniform radiation diagrams were positioned within the observed area and their parameters were adjusted (power, direction, position) taking into account electromagnetic field constraints in areas where people reside. Using here proposed GA aided method for transmitters’ parameter adjustment promising results were achieved.
|
M. Ponjavic (Građevinski fakultet u Sarajevu, Sarajevo, Bosnia and Herzegovina), A. Karabegovic (Faculty of Civil Engineering, Department of Geodesy, Sarajevo, Bosnia and Herzegovina)
Pareto-Based Genetic Algorithm in Multi-Objective Geospatial Analysis
This work studies the application of the multi-objective genetic algorithm based on the Pareto approach, as a tool for the decision making support in the geospatial analysis. Pareto-based evolutionary mechanism developed as an approach to multi-objective geospatial optimisation operates with fixed parameters of genetic operators. It can be used as efficient tool for multi-objective planning both for their power and flexibility and the fact that they generate a whole set of good solutions rather than just one "optimal" solution. Within the studies it is tested and suggested an adaptive mechanism for mutation parameter based on the determinsitic approach. The application of the suggested multi-objective Pareto based genetic algorithm over selecteded location problems demonstrates its ability of the discovery of multiple compromise solutions in a real spatial problem domain.
|
M. Bjelica (Faculty of Technical Sciences, Novi Sad, Serbia), M. Bjelica, N. Teslić (Fakultet tehničkih nauka, Novi Sad, Serbia)
A Concept and Implementation of the Embeddable Home Controller
This paper proposes a concept of a home controller for home automation that is embeddable to existing consumer electronic devices in the household, such as TV sets and set-top boxes. The concept is investigated through implementation in a digital set-top box extended with units for both power line (X10) and wireless (Zigbee) communication to control lights and appliances. Main goal was to propose a generic software model that would be able to support the majority of field-proven protocols for automation and lighting. Additional important aspects, such as home behavior modeling, comprehensive graphical user interface (GUI) and economic use of resources were covered.
|
J. Križaj, V. Štruc, N. Pavešić (University of Ljubljana, Faculty of Electrical Engineering, Ljubljana, Slovenia)
Adaptation of SIFT Features for Face Recognition under Varying Illumination
Face recognition has important role in many practical applications, like surveillance and access control. In this paper, a new face recognition system is presented, based on Scale Invariant Feature Transform (SIFT), proposed by Lowe in [1]. The SIFT technique and its corresponding SIFT features have many properties that make them suitable for matching different images of an object or a scene. The features are invariant to image scaling and rotation, (partial) occlusion and to certain extent also to changes in illumination.
SIFT algorithm has been used in many general object detection and recognition applications, and recently it was also applied to face recognition, where it was shown to deliver encouraging results. SIFT-based face recognition techniques found in the literature rely heavily on the so-called keypoint detector, which locates interest points in the given image that are ultimately used to compute the SIFT descriptors. While these descriptors are known to be among others (partially) invariant to illumination changes, the keypoint detector is not. Since varying illumination is one of the main issues affecting the performance of face recognition systems, the keypoint detector represents the main source of errors in face recognition systems relying on SIFT features.
To overcome the presented shortcoming of SIFT-based methods, a novel face recognition technique is proposed in this paper, that computes the SIFT descriptors at fixed points in the locations of the nodes on a regular grid, overlapping face image, during the training stage. By doing so, it eliminates the need for keypoint detection on the test images and renders our approach more robust to illumination changes than related approaches from the literature. The performance of the proposed approach was compared to the performance of some other face recognition techniques such as Principal Component Analysis (PCA) [2], Linear Discriminant Analysis (LDA) [3], and to several different modifications of the SIFT algorithm. Experimental results, obtained on the Extended Yale face database B [4], demonstrate that better results are achieved with proposed technique in comparison with the recognition performance of the remaining techniques assessed in our experiments, especialy under severe illumination conditions.
[1]. D.G. Lowe: Distinctive Image Features From Scale-Invariant Keypoints., International Journal of Computer Vision, Vol. 60, pp. 91–110, 2004.
[2]. M.Turk, A. Pentland: Face recognition using eigenfaces. IEEE Computer Society Conference on In Computer Vision and Pattern Recognition, Proceedings CVPR '91, pp. 586-591, 1991.
[3]. K. Etemad, R. Chellappa: Discriminant Analysis for Recognition of Human Face Images, Journal of the Optical Society of America, Vol. 14, pp. 1724-1733, 1997.
[4]. The Extended Yale Face Database B: http://vision.ucsd.edu/~leekc/ExtYaleDatabase/ExtYaleB.html.
|
Z. Đukić, Z. Đukić, Z. Kalafatić (Fakultet elektrotehnike i računarstva, Zagreb, Croatia), I. Otočan (PMF - Matematički odjel, Zagreb, Croatia)
An Experiment with Saccadic Eye Movement Simulation
Since human visual system surpasses any present artificial system in visual information processing, it is reasonable to investigate its properties. We explored and simulated saccade, biological mechanism used to get the most informative scene region into the central part of retina for high-level detail processing. Paper describes our implementation of saccade simulation system and discusses its applicability in image comparison and retrieval (query-by-example method) by examining properties of generated artificial eye-movement trajectories.
|
| Break |
| G2 |
T. Justin, R. Gajšek, V. Štruc, S. Dobrišek (Faculty of Electrical Engineering, University of Ljubljana, Slovenia, Ljubljana, Slovenia)
Comparison of different classification methods for emotion recognition
Automatic recognition of emotional states aims at automatically identifying the emotional or physical state of a human being from his or her voice. This complex problem has been researched by many scientists with the first papers on this topic appearing the mid-1980s. The hardest tasks in the field of automatic emotion recognition are divided in to three main parts; namely finding appropriate datasets, feature extraction, and classification. Finding suitable features from speech containing appropriate information about emotions states of speaker is still the major agitation for scientists around the world.
Features are normally extracted from acoustic data, which is included in emotional speech databases. Usually the databases differ in the language spoken, the number of the subjects, other physiological signals possibly recorded simultaneously with speech, the emotional states recorded, and the kind of the emotion featured in the database [1].
For the successful implementation of an emotional state classifier, good labeling of each emotional state is essential. It allows us to obtain sample features which bear sufficient information about the speaker's emotional state.
Only a carefully selected set of voice recordings can yield a group of attributes sufficiently rich with information. However, even more important is the method used for obtaining the features. A constrained version of Maximum Likelihood Linear Regression (CMLLR) [2] transformation of the monophone acoustic model, which was build using the Voicetran database [3] is one of the promising new approaches to the features extraction in the field of emotional state classification and is also considered here [4].
In this paper, we discuss classification features obtained through the CMLLR transformation of the acoustical speech model and the recorded voice segments of natural emotional states, which are part of the multi-modal AvID (audio-video) database [5]. The Weka machine learning workbranch is used to test the different classification algorithms to classify the MLLR features into one of two classes: the excited emotional state class and the normal emotional state class. Using a fixed and predefined experimental setup, several state-of-the-art classifiers are used, e.g. multilayer perceptron, decision trees, distance based classifiers and probabilistic classifiers. The highest average recall achieved is 70%. The results are presented in terms of the average recall and in the form of ROC [6] curves.
[1] Ververidis D. and Kotropoulos C.: Emotional speech recognition: Resources, features, and methods, Speech Communication 48 (2006), 1162–1181
[2] Gales M.J.F.: Maximum likelihood linear transformations for hmm-based speech recognition. Computer Speech and Language, 14 (1998), 75-98
[3] Mihelič, F., et al.: Spoken language resources at LUKS of the University of Ljubljana. Int. J. of Speech Technology, vol. 6 (3), pp. 221{232 (2006)
[4] R. Gajšek, V. Štruc, S. Dobrišek, F. Mihelič, Combining Audio and Video for Detection of Spontaneous Emotions, Proc. Interspeech 2009, Brighton, GB, ISCA, 2009, 1967-1970
[5] R. Gajšek, V. Štruc, F. Mihelič, Multi-Modal Emotional Database: AvID, journal Informatica, 33 (2009), 101-106
[6] T. Fawcett, An introduction to ROC analysis, Pattern Recognition Letters 27 (2006), 861-874
|
D. Lamovsky (, , ), R. Sadykhov (Belarusian State University of Informatics and Radioelectronics , Minsk, Belarus)
METHOD OF PEDESTRIANS TRAFFIC ASSESSMENT BASED ON ANALYSIS OF VIDEO DATA SURVEILLANCE SYSTEMS
This work represent a novel approach for people counting in video surveillance systems. In order to avoid the necessity of objects of interest detection, we propose to evaluate the number of passengers based on visible motion flow. It means that we consider the motion pedestrians not as the motion of individual objects, but as the movement of some mass (flow). This flow is estimated by the methods of background segmentation or by the methods of optical flow Error estimates of passengers using the technique used does not exceed 11%. The accuracy of the proposed method is considerably higher than the accuracy of methods based on tracking objects, especially if there are visual objects overlapping.
|
D. Bushenko (Belarusian State University of Informatics and Radioelectronics, Minsk, Belarus), R. Sadykhov (BSUIR, Minsk, Belarus)
Classification of Low-Contrast Extended Objects with Application to Textile Fibers Identification
Nowadays there is a demand of intelligent systems which are able to perform classification of color low-contrast extended objects such as textile fibers. Since this is a complex problem with no common way to solve it, detailed investigations are needed here. The paper presents a new way to identify the extended objects such as textile fibers. The workflow consists of three main steps: object extraction, splitting crossed objects and the object identification; and the algorithms for each step are presented. The most complex problem here is the crossed objects splitting which is a subtask of a general clustering task. The paper presents a new criteria of the selected object assessment which allows splitting the crossed objects using well-known optimisation techniques such as genetic algorithms. Finally, the paper presents the efficiency estimation of the proposed method of the extended objects identification and discusses the possible ways of further enhancement of the mentioned algorithms.
|
M. Katona, E. Neborovski, V. Marinković, M. Katona (Faculty of Technical Sciences, Novi Sad, Serbia)
Video quality assessment approach with field programmable gate arrays
The appearance of high definition digital content provides more impressive visual sensation by offering higher level of details compared to standard definition. At the same time, the amount of data to be transmitted and processed is significantly increased. In order to reduce bandwidth requirements in broadcasting systems, digital compression is applied.
Modern compression algorithms (MPEG-2, H.264) introduce artifacts in the video content. Depending on the compression factor and content characteristics, the artifacts are more or less noticeable, and significantly influence the subjective impression of quality. In order to maintain QoS in broadcast systems, the quality of video material has to be checked, in real-time for live streams or even faster for offline content.
The system presented in this paper is able to capture and measure video content of HD quality in real-time. It is developed as hardware architecture for a field programmable gate arrays - FPGA. At the moment, the system considers field-offset, blocking, ringing and blurring artifacts, but it can be easily extended with additional measures. The quality indicators are shown on remote monitoring workstation, making possible for remote operator to judge whether the video quality is satisfactory or not.
The reference algorithms for video quality assessment are written in C++ language, and represent functional models. These models are updated for fixed point precision suitable for hardware implementation. Verilog hardware description language is used to describe RTL model for FPGA. Verification is done on two levels. Firstly, with cosimulation between fixed-point algorithm model in C++ and RTL model and secondly by post place&route simulation in order to assure that there are no issues in the FPGA synthesis process. For real-time testing the system is integrated within video testing environment with capability to record full HD video stream. Video data is simultaneously stored in onboard DDR2 memory and analyzed with implemented algorithms. Grabbed video data is transferred to a workstation for off-line analysis with reference algorithm functional models and results are compared with captured real-time data.
The novelty of this paper is in demonstration of hardware architecture for real-time video quality assessment of the HD video content and comparison with similar software solutions with multicore data processing units.
References:
[1]. Z. Wang; A. Bovik, B. Evan, “Blind measurement of blocking artifacts in images”, International Conference on Image Processing, pp. 981 - 984 vol.3, 2000
[2]. X. Feng,J. P. Allebach, “Measurement of Ringing Artifacts in JPEG Images”, Proceedings of the SPIE, Volume 6076, pp. 74-83, 2006
[3]. H. Hu, G. de Haan, “Low cost robust blur estimator”, IEEE International Conference on Image Processing, pp. 617-620, 2006
[4]. N. Lukic, I. Papp, Z. Marceta, D. Acanski and M.Temerinac, “Evaluation of CELL IBM platform regarding development of advanced real-time video algorithms”, IEEE ICEST 2008, June 2008
[5]. I. Papp, N. Lukic, Z. Marceta, N. Teslic, M. Schu, "Real-time video quality assessment platform," icce, pp.1-2, 2009 Digest of Technical Papers International Conference on Consumer Electronics, 2009
[6]. D. Marijan, N. Teslic, M. Temerinac, V. Pekovic: "On the effectiveness of the system validation based on the black box testing”, IEEE Circuits and Systems International Conference on Testing and Diagnosis, 2009
|
I. Paukner Stojkov, S. Lončarić (FER, Zagreb, Croatia)
A method for eye detection in face images
Eye detection in images is an important part of human face segmentation. This paper proposes a two step solution. The first step applies Bayesian classifiers and connected components labeling for determining the face position. In detection process all of the components of the selected color space are used. The second step includes a method that applies a corner detector and histograms of vertical and horizontal projections of detected corners. The eye position is detected by using several iterations of the second step. The tested method shows good results in images like those used for personal identification.
|
| Lunch break |
| G3 |
D. Brodic (University of Belgrade, Bor, Serbia), B. Dokic (Faculty of Electrical Engineering, Banja Luka, Bosnia and Herzegovina)
Reference Text Line Identification Using Gaussian Kernel Extended by Morphological Operations
In this paper, new approach to reference text line identification by Gaussian kernel is presented. As a result of algorithm, defined area exploited for text segmentation. Using morphological operations reference text line is estimated. Method is examined and evaluated under different printed and “handwritten” text skew angles. Results are examined, analyzed and discussed. Proposed algorithm showed robustness for different types of skewness
|
A. Bulović, D. Bučar, P. Palašek, B. Popović, A. Trbojević, L. Zadrija, K. Brkić, Z. Kalafatić, S. Šegvić (Fakultet elektrotehnike i računarstva, Zagreb, Croatia)
Streamlining Collection of Training Samples for Object Detection and Classification in Video
This paper is concerned with the field of object recognition and detection in computer vision. Many promising approaches in the field exploit implicit knowledge contained in groundtruth represented by a collection of manually annotated object appearances. In the resulting paradigm, the recognition algorithm is automatically constructed by some machine learning technique. It has been shown that the quantity and quality of positive and negative groundtruth samples is critical for good performance of such approaches.
However, collecting the groundtruth requires tedious manual effort which is expensive in time and prone to error. In this paper we present design and implementation of a software system which addresses these problems. The system features a bootstrapping approach whereby the current state-of-the-art detection and recognition algorithms are used to streamline the collection of additional groundtruth. The architecture of the proposed system allows to seamlessly connect the main GUI component in Java with computationally demanding components written in a compiled language such as C. The system has been designed in a modular manner in order to promote testability and reuse of the existing components. Finally, we also discuss implementation details which were were found to be critical for the usability of the final product.
|
S. Stancescu (University POLITEHNICA Bucharest, Bucharest, Romania), C. Ingineru (University Politechnica Bucharest, Bucharest, Romania)
Toward Application Driven and Included Compiler
Each application has proprietary data structures and processing rules. By developing applications through HLL compilers, source files, written in a HLL, are an intermediate representation stage interposed between user and developer. Instead, a different developer paradigm is proposed, by choosing a specific user-friendly language, specially designed for each application, and by including in the application the appropriate compiler. An example is provided, as a Web editor. The application development includes a universal procedure that grows gradually with the included compiler. In general lines, the paradigm is based on three tiers, each of them being represented or managed by one type of characters: the Architect who provides general guidelines, the Developer who adds more concrete information and the User who use the system. The Developer creates concrete data structures by providing procedural treatment, which can be shared with others Developers or Users on the communication media - Internet. Applying syntactic rules, also named parsing, we obtain an intermediary and very general result, which will be personalized by making it more concrete and closest to a specific field of interest.
|
D. Marčetić (Fakultet elektrotehnike i računarstva, Zagreb, Croatia), S. Ribarić, D. Marčetić (akultet elektrotehnike i računarstva, Zagreb, Croatia)
An Associative Information Retrieval Algorithm for a Kanerva-like Memory Model
Abstract-This paper presents an associative information retrieval algorithm for a Kanerva-like sparse distributed memory (SDM) model. This memory model is used to implement the associative level of a hierarchical heterogeneous knowledge-base model consisting of multi-levels, starting from an associative level, through to the semantic, rule-based and description-generator level as the top level in the hierarchy. The architecture of knowledge-base was inspired by biological and psychological models. The proposed algorithm retrieves concepts from the associative level based on the similarity between a concept of interest and already stored concepts. The similarity is expressed by a value of the linguistic variable. With this approach it is possible to solve a problem when the inference processes at the semantic level encounter an unknown concept of interest. The algorithm is demonstrated by retrieving concepts that were stored based on the results of psychological experiment.
|
K. Križanović, Z. Galić, M. Baranović (FER, Zagreb, Croatia)
Spatio-Temporal Data Streams: An Approach to Managing Moving Objects
Spatio-temporal database is a new type of database that manages spatio-temporal objects and supports corresponding query languages. If position and/or extent of an objects changes continuously, we talk about moving object database. DSMS (Data Stream Management System) is an extension of a DBMS (Database Management System) that efficiently manages potentially unbound data streams as well as standard relations and supports continuous queries in addition to standard one-time queries. By combining these two systems, we get a system that manages moving objects whose position is determined by one or more data streams. In this paper we present a data model consisting of data types and operations needed to support geospatial and spatio-temporal data in data streams. Defined data model is illustrated through an SQL-like query language.
|
V. Sesar, I. Mišković, T. Šijak (BRODARSKI INSTITUT, Zagreb, Croatia)
Use of Describing Functions in Analysis of Systems with Neural Networks and Dominant Nonlinearities
The application of intelligent controllers such as neural networks and fuzzy controllers in technical systems is increasing. The main reason for this is that exact mathematical model parameters of the controlled process are not essential. On the other hand, a great downside to this type of control is that stability cannot be proved.
Technical systems often exhibit nonlinear behavior which can be described with elementary nonlinear elements such as hysteresis and/or dead zones. This paper presents the use of the describing function as a means of analysis of systems with neural networks and dominant nonlinearities. The equivalent describing function can then be used for stability analysis using, for example, the Popov criterion.
The nonlinear elements used in this paper are three position relay with hysteresis and saturation with dead zone. Equivalent describing functions of the nonlinearities in combination with neural networks have been obtained experimentally, using the method based on piecewise affine linearization of the static characteristic. A comparison of the obtained describing functions is presented. In addition to that, the time responses of the equivalent and original systems are given, proving the applicability of the method.
|
| Break |
| G4 |
I. Sikirić (Mireo d.d., Zagreb, Croatia), K. Brkić (Fakultet elektrotehnike i računarstva, Zagreb, Croatia), S. Šegvić (Fakultet elektrotehnkie i računarstva, Zagreb, Croatia)
Recovering a Comprehensive Road Appearance Mosaic from Video
We describe a system that employs a single
calibrated camera mounted on a moving vehicle to produce
a 2D road map as a comprehensive mosaic of individual
orthogonal views. The system first transforms the image of
the road acquired from a driver’s perspective into an orthogonal
view by inverse perspective mapping. Consequently, the
current orthogonal image is aligned with the previous one by
an exhaustive search optimization technique. Experiments
have been performed on videos taken along public roads
through Croatian countryside and small cities. The obtained
results are provided and discussed.
|
A. Martinović, G. Glavaš, M. Juribašić, D. Sutić, Z. Kalafatić (Fakultet elektrotehnike i računarstva, Zagreb, Croatia)
Real-time Detection and Recognition of Traffic Signs
Automated recognition of traffic signs is becoming a very interesting area in computer vision with clear possibilities of its application in automotive industry. For example, it would be possible to design a system which could recognize the current speed limit on the road and notify the driver in an appropriate manner. In this paper we deal with methods for automated localization of certain traffic signs, and classification of those signs according to official designations. We propose two different approaches of determining the current speed limit after the sign was localized. A demo software system was developed to demonstrate the presented methods. Finally, we compare results obtained from the developed software, and discuss the influence of different parameters on recognition performance and quality.
|
L. Načinović, L. Načinović, S. Martinčić-Ipšić, I. Ipšić (Odjel za informatiku, Sveučilište u Rijeci, Rijeka, Croatia)
Intonation Modeling for Croatian Speech Synthesis
Intonation modelling has become a very important part of speech synthesis and speech recognition systems. This paper summerizes basic concepts of prosody and gives an outline of intonation function. Most important applications of intonation and forms of intonation with examples from Croatian language are presented. It also outlines various prosody modelling approaches, labelling systems and approaches to the generation of prosodic information. Goals, significance and benefits of intonation research on Croatian language in the upcoming years are discussed.
|
M. Klarić (DALEKOVOD d.d., ZAGREB, Croatia)
Predviđanje vremenskog slijeda mjernih podataka vjetra uporabom kaskadno-korelacijskih neuronskih mreža
Stohastičke veličine koje se mjere u vremenskom slijedu i čiji se krajnji rezultati izražavaju u vremenskim prosjecima, u velikom broju slučajeva pokazuju statističku
pravilnost pojavljivanja. Primjeri takvih veličina jesu primjerice brzina i smjer vjetra, ali i ostale atmosferske veličine. U ovom radu predstavljena je jedna metoda za procjenu vremenskog slijeda stohastičkih mjernih veličina na temelju mjerenja karakteristika vjetra na udaljenim mjernim postajama. Uporabljena je metoda kaskadno-korelacijskih neuronskih mreža s algoritmom brze propagacije za učenje mreže te tangensom hiperbolnim kao aktivacijskom funkcijom.
|
Ž. Deljac (HT - HRVATSKE TELEKOMUNIKACIJE d.d., ZAGREB, Croatia), R. Moštak, T. Stjepanović (HT-Croatian telecommunications, Zagreb, Croatia)
The use of Bayesian networks in recognition of faults causes in the BB networks
The aim of this paper is to explore possibilities of using Bayesian networks in recognizing and finding places and causes of faults and detection of degradation of service in broadband telecommunications networks. To improve the functionality of the BB network operators often maintain a variety of regular or emergency work (upgrade, replace equipment, add new functions, ..) on the various elements of the network that can have caused the degradation of services. This degradation in some cases is not recognized immediately by the system alarms, but later appears in the form of random disturbances reported by the users of these services. This paper explores how to use Bayesian networks in the separation of these faults from the many other "regular disturbances" that were reported by users.
|
I. Ninčević, M. Jadrić, Ž. Garača (Faculty of Economics, Split, Croatia)
Simulating and Optimizing a Core Business Process in a Sales Company
The aim of this research is to design a better and more efficient process of sale and delivery of pipes in a sales company in Split, Croatia. For this purpose, a model based on the current performance of this core process (AS-IS), will be shown. The purpose is to establish such business process where there would be no queuing tails and where the resources are used efficiently. Eliminating these problems would then reduce operational costs. The optimization and simulation of the process seek to find an optimal (TO-BE) solution. The processes are modeled in ARIS Business Arhitect 7.0 while optimization is done with OptQuest from simulation tool Arena 7.0.
|
|
Basic information:
Chairs:
Slobodan Ribarić (Croatia), Andrea Budin (Croatia)
International Program Committee Chairman:
Petar Biljanović (Croatia)
International Program Committe:
Alberto Abello Gamazo (Spain), Slavko Amon (Slovenia), Michael E. Auer (Austria), Mirta Baranović (Croatia), Ladjel Bellatreche (France), Nikola Bogunović (Croatia), Peter Brezany (Austria), Željko Butković (Croatia), Željka Car (Croatia), Matjaž Colnarič (Slovenia), Alfredo Cuzzocrea (Italy), Marina Čičin-Šain (Croatia), Dragan Čišić (Croatia), Todd Eavis (Canada), Maurizio Ferrari (Italy), Jasna Glavaš (Croatia), Matteo Golfarelli (Italy), Stjepan Golubić (Croatia), Francesco Gregoretti (Italy), Niko Guid (Slovenia), Yike Guo (United Kingdom), Ladislav Hluchy (Slovakia), Vlasta Hudek (Croatia), Željko Hutinski (Croatia), Mile Ivanda (Croatia), Robert Jones (Switzerland), Peter Kacsuk (Hungary), Aneta Karaivanova (Bulgaria), Miroslav Karasek (Czech Republic), Bernhard Katzy (Germany), Christian Kittl (Austria), Miljenko Krvišek (Croatia), Mladen Mauher (Croatia), Branko Mikac (Croatia), A. Min Tjoa (Austria), Jadranko F. Novak (Croatia), Jesus Pardillo (Spain), Nikola Pavešić (Slovenia), Ivan Petrović (Croatia), Radivoje S. Popović (Switzerland), Ognjen Prnjat (Greece), Slobodan Ribarić (Croatia), Karolj Skala (Croatia), Ivanka Sluganović (Croatia), Vanja Smokvina (Croatia), Ninoslav Stojadinović (Serbia), Aleksandar Szabo (Croatia), Laszlo Szirmay-Kalos (Hungary), Jadranka Šunde (Australia), Antonio Teixeira (Portugal), Roman Trobec (Slovenia), Ivana Turčić Prstačić (Croatia), Walter Ukovich (Italy), Mirko Varga (Croatia), Boris Vrdoljak (Croatia), Dalibor Vrsalović (Croatia), Robert Wrembel (Poland), Baldomir Zajc (Slovenia)
Scope:
The conference addresses topics related to underlying technologies of intelligent systems, methods and tools, as well as all kinds of applications.
- Machine Intelligence
- Human-Machine Interfaces
- Computer Vision
- Image Processing
- Pattern Recognition
- Speech Recognition and Synthesis
- Soft Computing
- Expert Systems
- Data Mining
- Data Warehousing
- Machine Learning
- Knowledge Management – Methods & Tools
- Business Intelligence
- Multi-Agent Systems
- Biometrics
- Computer Architectures for Intelligent Systems
Official languages are English and Croatian.
Location:
Opatija, often called the Nice of the Adriatic, is one of the most popular tourist resorts in Croatia and a place with the longest tourist tradition on the eastern part of Adriatic coast. Opatija is so attractive that at the end of the 19th and beginning of the 20th centuries it was visited by the most prominent personalities: Giacomo Puccini, Pietro Mascagni, A. P. Čehov, James Joyce, Isidora Duncan, Beniamino Gigli, Primo Carnera, Emperor Franz Joseph, German Emperor Wilhelm II, Swedish Royal Couple Oscar and Sophia, King George of Greece.
The offer includes 20-odd hotels, a large number of catering establishments, sports and recreational facilities.
For more details please look at www.opatija.hr/ and www.opatija-tourism.hr/.
Registration / Fees:
|
REGISTRATION / FEES
|
Price in EUR
|
|
Before May 09, 2011
|
After May 09, 2011
|
| Members of MIPRO and IEEE |
180
|
200
|
| Students (undergraduate), primary and secondary school teachers |
100
|
110
|
| Others |
200
|
220
|
Contact:
Slobodan Ribarić
Faculty of Electrical Engineering and Computing
Unska 3
HR-10000 Zagreb, Croatia
Phone: +385 1 612 99 52
Fax: +385 1 612 96 53
E-mail: slobodan.ribaric@fer.hr
Andrea Budin
Ericsson Nikola Tesla Inc.
R&D Center
Krapinska 45
HR-10000 Zagreb, Croatia
Phone:+385 1 365 34 23
Fax: +385 1 365 35 48
E-mail: andrea.budin@ericsson.com
|
|
|
|
| Currently there are no news |
|
|
|
|
|