Search  English (United States) Hrvatski (Hrvatska)

innovative promotional partnershipArtificial Intelligence towards EU Multilingualism

Technical co-sponsorship

 
The authors (except for invited speakers) are kindly asked to keep their presentations within the timeframe of approximately ten (10) minutes.

Event program
Thursday, 5/29/2014 9:30 AM - 10:30 AM,
Leut I, Hotel Admiral, Opatija
Invited speech for BiForD and CIS 
A. K. Jain (Michigan State University, USA)
Biometric Recognition: Some Challenges in Forensics 
Thursday, 5/29/2014 11:00 AM - 1:00 PM,
Nava 2, Hotel Admiral, Opatija
Papers 
1.J. Kaljun, G. Harih, B. Dolšak (University of Maribor, Faculty of Mechanical Engineering, Maribor, Slovenia)
Intelligent Support Used for Providing a Pleasant User Experience 
Desire is the fundamental motivation of all human action (Thomas Hobbes, 1588 – 1679). When using a “desired” product, user “feels” “pleasure” and “comfort” and it can be said that user has a “pleasant user experience”. Today desire is the main marketing weapon. Stimulating a desire for a certain product by conventional advertising has to be supported by products’ “self-expression”. With the growth and globalization of the market, the future users of the specific product, in this case consumer products (household appliances, hand tools…), increasingly search, choose and buy desired product in virtual environment e.g. online stores. In such a case, when selecting the product, users rely on: (a) a description of the product or. technical system, which is usually prepared by the manufacturer, (b) the comments of existing users, (c) test reports carried out by the relevant institutions and (d) virtual product presentation. It is a virtual visual perception of the product that is crucial, as it must support the aforementioned descriptions. Products’ form has to reflect the correct combination of, for the user, important characteristics such as functionality, ergonomics, practicality, robustness, elegance... In the virtual world a visual perception of products’ form is a point of contact between the product and the prospective users. On the other hand also in the real world, the prospective user, when selecting a product, among wide range of products of same category located on the shelf, primary chooses that or those, whose form most clearly express the aforementioned properties, which are important to the user. The modern process of designing technical systems, including consumer products, specifically household appliances and hand tools, is a relatively well-defined and relatively well-supported by modern computer tools. Support offered by many commercial and specially developed computer tools, mainly refers to the graphic representation, simulation, engineering analysis (e.g. finite element method) and the animation of the future operation of the technical system in a virtual world. In fact, these tools do not affect the scope of defining the form or design. Despite the wide range of possibilities provided by the modern computer tools, the designer has to evaluate the information, find interdependences and end up with a broad list of possible partial solutions, among which an optimum combination has to be selected. Alongside this, the designer has to take care of a very complex and to a certain degree "subjective" area of aesthetic products. For this purpose, the designer has to know all the spheres of influence, their possible overlapping or exclusions and of course all of the available design (visual) elements. With its true/false decision seriously affects the attractiveness and usefulness of the product in practice, as well as technical feasibility, lifetime, economic viability, recyclability... In taking such decisions many different experts from various areas of expertise should cooperate, which often results in difficulties in coordinating the decisions that need to be in line with sustainable development. The paper will discuss a framework of intelligent support system for the definition of the relevant visual perception of the product on the basis of functional, ergonomic and technological criteria, which will be able to offer support to the designer in the decision making process of selecting the most appropriate aesthetic (design) characteristics - features for a specific product or technical system.
2.D. Helic (TU Graz, Graz, Austria)
Regular Equivalence in Informed Network Search 
Search in networks is defined as a process in which an agent hops from one network node to another by traversing network links in search for given nodes. The simplest example of network search is the random walk where the agent at any given step selects one link uniformly at random from all outgoing links of the current node. Contrary to the random walk in informed search the agent possesses background knowledge of the network which steers its decisions when selecting the next link to traverse. For example, in decentralized search the background knowledge of the network is represented as a distance matrix with distances between all pairs of nodes in the network. In its simplest form decentralized search is greedy and deterministic, that is an agent always select as the next node to hop to a node which has the smallest distance to a given target node. The distance matrix can be calculated in various ways in order to model various search scenarios or to best fit needs of an application. For example, distances based on node similarities or node degrees have been commonly used in the past. In this paper we evaluate the measures that capture regular equivalence of nodes in a network in respect to their suitability as a distance metric to inform search in networks. In particular we are interested in the properties of Katz similarity for this task.
3.D. Domović, T. Rolich, D. Grundler, S. Bogović (Tekstilno-tehnološki fakultet, Zagreb, Croatia)
Algorithms for 2D Nesting Problem Based on the No-Fit Polygon 
Strip-packing problem is apparent in textile industry where a set of items, i.e. cutting parts (2D convex or non-convex polygons) need to be placed on a rectangular container (fabric with an m×n area) so that cutting parts do not overlap and do not exceed the boundaries of the container. The goal is to find a placement that utilizes the area of a container. In this paper three methods (random search, greedy algorithm and genetic algorithm) are tested on sets of regular (convex polygons) and irregular (cutting parts) items. The goal is to find the optimal items placement that minimizes the cover area. In this paper a no-fit polygon (NFP) is used to assure two items touch without overlapping. NFP is constructed by rotating polygon B around a static polygon A in a way their edges always touch and never overlap. The result is a polygonal area enclosed by trajectory of rotating polygon's reference point which represents the overlapping area of A and B. Items touch if polygon B is placed on the NFP’s border. Non-convex cutting parts are approximated with their convex hull since a NFP version for convex polygons is used in this paper.
4.E. Cherkashin, S. Badmatsyrenova (Institute of System Dynamics and Control Theory SB RAS, Irkutsk, Russian Federation), I. Vladimirov (V.B.Sochava's Institute of Geography SB RAS, Irkutsk, Russian Federation), A. Popova, A. Davydov (Institute of System Dynamics and Control Theory SB RAS, Irkutsk, Russian Federation)
An Optimal Control Module for Sustainable Natural Resources Consumption Control Synthesis for Decision Support Systems 
We consider optimal control problems solved in the process of decision support implemented as submodules of a hybrid decision support systems (HDSS) for sustainable consumption of the forest resources of an industrial region. The main cycle of the decision support consists of data acquisition, model construction of the natural resource; model investigation, e.g., prognosis computer simulation; multicriterial result assessment, and the resulting data visualization. The initial data are obtained from the databases representing an initial state of the object under investigation, generated scenario variant constraints, user's answers for the refinement questions. The model construction is carried on the base of the logical inference of a set of subgoals over the input data, using a knowledge base. Next step is to investigate and use the constructed model to obtain data for counterintuitive decision. The paper, in general, is devoted to this problem. The multicriterial data assessment is carried out on the base of Pareto set construction and its reduction with special human expert-driven procedures. Data visualization is based on construction of diagrams and cartographical layers with free and open-source software. The cycle of the model investigation consists of tests of functional properties to be, e.g., continuous and differentiable on the modelling interval, availability of the necessary derivatives, etc. Next step is the optimal control problem solution satisfying the main substantional criterium of the predefined criteria set, which is chosen according to user's preference. The results of the solution are stored in a database for further usage. The model is constructed in form of tagged formulae. The data structures can be used to produce the formulae of the derivatives, calculate vector and matrix time series, as well as describe operations in a meaningful terms of the object properties to allow a dialog subsystem to carry on user interaction to visualize data. An example of the HDSS functioning is considered for the problem of Irkutskaya Oblast' (Irkutsk Region) forestries felling prognosis.
5.S. Picek (Radboud University, Nijmegen, Croatia), D. Jakobović (Fakultet elektrotehnike i računarstva, Zagreb, Croatia), R. Boix Carpi (Riscure BV, XJ Delft , Netherlands), L. Batina (Radboud University Nijmegen, HC Nijmegen , Netherlands)
Evolving Genetic Algorithms for Fault Injection Attacks 
Genetic algorithms are used today to solve numerous difficult problems. However, it is often needed to specialize and adapt them further in order to successfully tackle a specific problem. One such example is the fault injection attack, where the goal is to find a specific set of parameters that can lead to a successful cryptographic attack in a minimum amount of time. In this paper we address the process of the specialization of genetic algorithm from its standard form to the final, highly-specialized one. In this process we customize variation operators, add mapping between the values in cryptograpic domain and genetic algorithm domain and adapt genetic algorithm to work on-the-fly. In the last phase of development we go to the memetic algorithm by adding a local search. Furthermore, we give a comparison between our algorithm and random search which is the usual method of choice at the moment. The experiments show that our approach significantly outperforms the random search .
6.A. Jović, K. Brkić, N. Bogunović (Fakultet elektrotehnike i računarstva, Zagreb, Croatia)
An Overview of Free Software Tools for General Data Mining 
This expert paper describes the characteristics of six most used free software tools for general data mining that are available today: RapidMiner, R, Weka, KNIME, Orange, and scikit-learn. The goal is to provide the interested researcher with all the important pros and cons regarding the use of a particular tool. A comparison of the implemented algorithms covering all areas of data mining (classification, regression, clustering, associative rules, feature selection, evaluation criteria, visualization, etc.) is provided. In addition, the tools' support for the more advanced and specialized research topics (big data, data streams, text mining, etc.) is outlined, where applicable. The tools are also compared with respect to the community support, based on the available sources. This multidimensional overview in the form of expert paper on data mining tools emphasizes the quality of RapidMiner, R, Weka, and KNIME platforms, but also acknowledges the significant advancements made in the other tools.
7.I. Markić (Croatia osiguranje d.d., Ljubuški, Bosnia and Herzegovina), M. Štula, Split, Croatia), J. Maras (Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture (FESB), Split, Croatia)
Intelligent Multi Agent Systems for Decision Support in Insurance Industry 
MAS (Multi Agent Systems), belong to the field of AI (Artificial Intelligence), the study addressing the methods of construction of complex systems using a large number of agents and adjusting their behavior to a particular problem. MAS can be defined as groups of entities (agents) working together to solve problems not solvable by their individual skills and know-how. Business decisions are based on extraction of useful knowledge from different data sources. Those data sources, usually called the big data, might be huge digital data sets such as an internal data warehouse or external sources like the web. In order to process the big data, one needs to have smart solutions, analytics services, programming tools and applications. Intelligent agents can perform certain tasks such as big data processing, information retrieval and the creation of OLAP (On-Line Analytical Processing) cubes which represent multidimensional data. Given that volumes of data grow continuously, insurance companies have to be able manage them. The agent technology has the potential to become the key technology for management of ever-increasing amount of information. This paper describes how to develop MAS in order to gather useful knowledge from the data source and automate the specific process for better business decision.
8.K. Sylejmani (Department of Computer Engineering, University of Prishtina, Prishtina, Kosovo), A. Muhaxhiri (ProSolution, Prishtina, Kosovo), A. Dika, L. Ahmedi (Department of Computer Engineering, University of Prishtina, Prishtina, Kosovo)
Solving Tourist Trip Planning Problem by Using Simulated Annealing Algorithm 
Nowadays, by help of tourist trip planners, in many travel destinations, the trip itinerary could be prepared automatically. The trip planners enable customization of trip itinerary based on tourist preferences, available time and budget, where the itinerary is optimized by using optimization techniques from the field of metaheuristics. In this paper, we present a simulated annealing approach to solve the tourist trip planning problem. The hard constraints include limited trip duration, working hours of Points of Interest (POIs) and tourist budget, while tourist preferences and traveling time comprise the soft constraints, which take part in a double criterion function for evaluation of quality of solutions. The search space is explored by using three types of moves, namely Insert, Swap and Shake. The algorithm is tested against some real test data for the city of Prishtina, Kosova. In addition to fine tuning the values of algorithm parameters, further experiments are made for exploring different approaches of construction of initial solution, time duration of algorithm execution for various trip lengths, and influence of Shake move into the solution quality.
9.E. Mešković, D. Osmanović (University of Tuzla, Faculty of electrical enggineering, Tuzla, Bosnia and Herzegovina), Z. Galić, M. Baranović (University of Zagreb, Faculty of Electrical Engineering an Computing, Zagreb, Croatia)
Generating Spatio-Temporal Streaming Trajectories 
Recent research efforts in data stream management systems (DSMS) focus mainly on processing continuous queries over traditional data streams, and only a few addressed spatio-temporal continuous queries. OCEANUS presents an effort to extend TelegraphCQ DSMS with spatial support providing a platform for spatio-temporal streaming applications. Data type system that represents the formal basis for modeling moving objects in data streams, as well as an approach for managing moving objects in spatio-temporal data streams based on user-defined aggregate functions (UDAF) are presented. In this paper we are concerned with improving proposed approach for extracting moving objects out of spatio-temporal data streams based on UDAF. It is based on two methods for detecting static locations on trajectories of moving objects and for reducing total number of units in sliced representation by introducing trajectory buffer. First method leads to more precise results of certain operations over moving objects such as retrieving information about speed and direction, but also enables introduction of few new operations such as retrieving static locations and overall stoppage time on moving object's trajectory. Second method leads to better memory usage during processing spatio-temporal continuous queries in DSMS.
Thursday, 5/29/2014 3:00 PM - 7:00 PM,
Nava 2, Hotel Admiral, Opatija
Invited speech 
N. Nain (Department of Computer Engineering, MNIT Jaipur, India)
Indi Scripts Corpus Development and Handwritten Text Recognition 
Papers 
1.M. Čupić, K. Brkić, T. Hrkać, Ž. Mihajlović, Z. Kalafatić (FER, Zagreb, Croatia)
Automatic Recognition of Handwritten Corrections for Multiple-Choice Exam Answer Sheets 
Automated grading of multiple-choice exams is of great interest in university courses with a large number of students. We consider an existing system in which exams are automatically graded using simple answer sheets that are annotated by the student. A sheet consists of a series of circles representing possible answers. As annotation errors are possible, a student is permitted to alter the annotated answer by annotating the "error" circle and handwriting the letter of the correct answer next to the appropriate row. During the scanning process, if an annotated "error" circle is detected, the system raises an alarm and requires intervention from a human operator to determine which answer to consider valid. We propose a computer vision algorithm that enables automated reading of handwritten answers, which minimizes the need for human intervention in the scanning process. We test our algorithm on a large dataset of real scanned answer sheets, and report encouraging performance rates.
2.D. Brodic (Technical Faculty in Bor, Bor, Serbia), Z. Milivojevic (Technical College NIs, Nis, Serbia), C. Maluckov, M. Jevtic (Technical Faculty in Bor, Bor, Serbia)
Discrimination between Serbian and Slovenian Language by Texture Analysis 
The paper proposed the language identification method according to the co-occurrence analysis. First, each letter was encoded into the cipher. Such a cipher was subjected to a feature extraction. Then, the local binary pattern as well as its expanded version called adjacent local binary pattern was extracted. Because the differences in the language characteristics, the analysis showed diversity. This diversity established the discrimination criteria between tested languages. The proposed method was tested on a database that included documents in Serbian and Slovenian language. The experiments gave encouraging results.
3.D. Khanduja, N. Nain (MNIT Jaipur, Jaipur, India)
Script Independent Feature Set for Handwritten Text Recognition 
This paper proposes a hybrid approach combining the structural features like number of end points, loops and intersection points of the character and a mathematical model of curve fitting on statistically zoned partitions to simulate the best features of a character, forming a feature vector of coefficients of the curve. This vector is combined with the spatial distribution of the foreground pixels for each zone and hence script independent feature representation. The approach has been evaluated experimentally on English, Hindi and Urdu scripts achieving an average recognition accuracy of 80 – 85% for any script without incorporating any script specific features.
4.S. Panwar, N. Nain (Malaviya National Institute of Technology, Jaipur, India)
An Efficient Feature Extraction Method for Segmented Cursive Characters Recognition 
Handwritten document analysis is a persistent research area nowadays. It has drastic applications in various domains as human computer interaction, machine translation and automation of reading various scanned images for handwritten fields in forms, postal address on envelopes and amounts in banks checks. The main factor which influence the performance of handwritten text recognition is the selection of an appropriate set of features for representing input samples. In this paper, we describe the feature extraction phase of handwritten document recognition system and introduce a hybrid and efficient feature vector for each handwritten character. We have used the statistical and structural properties of the character to represent the particular character class. Experiments have been performed by using standard database of handwritten digit and standard database of handwritten letters. We have tested the accuracy of recognition with different classification methods. The results have been compared with existing features extraction algorithms. The comparative results shows the effectiveness of our approach.
5.P. Choudhary (National Institute of Technology Manipur, Imphal, India), N. Nain (Malaviya National Institute of Technology Jaipur, Jaipur, India)
An Annotated Urdu Corpus of Handwritten Text Image and Benchmarking of Corpus 
For a linguistics related research on a language there is always need a large collection of database which include all features of language such as grammatical information, style of writing etc. Corpus provide a platform for investigation on a natural language. As compared to other language very limited research work is done on Urdu language due to its segmented dilemma and difficult character shape. Very less number of printed text data available in Urdu most of data is available in graphical format. To increase Natural Language Processing research work on Urdu language there is need a large database which contain a large number of variance annotated Urdu handwritten as well as printed text. In our purposed work we design a large database of Urdu text including 1000 handwritten text image of 500 different writers. The estimation number of words would be around 3.5 million. Selection of words will be done from different category so that maximum number of distinct words can include. Corpus would be annotated for line as well as word where word may be an individual character or component, to provide facilities such as information extraction and detection.
6.S. Beliga, S. Martinčić-Ipšić (Sveučilište u Rijeci - Odjel za informatiku, Rijeka, Croatia)
Non-Standard Words as Features for Text Categorization 
This paper presents categorization of Croatian texts using Non-Standard Words (NSW) as features. Non-Standard Words are: numbers, dates, acronyms, abbreviations, currency, etc. NSWs in Croatian language are determined according to Croatian NSW taxonomy. For the purpose of this research, 390 text documents were collected and formed the SKIPEZ collection with 6 classes: official, literary, informative, popular, educational and scientific. Text categorization experiment was conducted on three different representations of the SKIPEZ collection: in the first representation, the frequencies of NSWs are used as features; in the second representation, the statistic measures of NSWs (variance, coefficient of variation, standard deviation, etc.) are used as features; while the third representation combines the first two feature sets. Naive Bayes, CN2, C4.5, kNN, Classification Trees and Random Forest algorithms were used in text categorization experiments. The best categorization results are achieved using the first feature set (NSW frequencies) with the categorization accuracy of 87%. This suggests that the NSWs should be considered as features in highly inflectional languages, such as Croatian. NSW based features reduce the dimensionality of the feature space without standard lemmatization procedures, and therefore the bag-of-NSWs should be considered for further Croatian texts categorization experiments.
7.I. Budiselić, G. Delač, K. Vladimir (FER, Zagreb, Croatia)
Developing a Text Classifier With Constrained Development and Execution Time 
The aim of this paper is to show that an accurate and efficient text classifier for relatively simple problem domains can be created in only a few hours of development time. The motivating example discussed in the paper is a recent HackerRank competition problem that tasked competitors with creating a classifier for questions from the popular question and answer platform StackExchange. The paper describes the key components of one solution to this problem, and briefly overviews the naive Bayes classifier that is the basis of the solution. The discussion is focused on feature selection and example representation which were the key challenges to be addressed during the development of this classifier. We also analyze the effect of the number of features on accuracy, training and classification time and the size of the resulting classifier and the representation of the training examples which were all important characteristics for the competition. The described classifier achieved slightly over 89% accuracy on the hidden question set, while the winning submission achieved around 92%.
8.M. Rauker Koch (Railway and Technical School Moravice, Moravice, Croatia), M. Pavlić (Department of Informatics, University of Rijeka, Rijeka, Croatia), A. Jakupović (Business Department, Polytechnic of Rijeka, Rijeka, Croatia)
Application of the NOK Method in Sentence Modeling 
Knowledge representation is one of the areas covered by artificial intelligence. This paper describes the method NOK (Nodes Of Knowledge), which allows a graphical representation and organization of knowledge in the form of a diagram as a network of nodes. Using this method, knowledge in textual form is transformed into a model that can be written to a database without changing the meaning of the sentence. Using sentences from Aesop's fable in Croatian („Zlatna jaja u kokoši“) and English ("The Goose with the golden eggs") versions as an example, the application of NOK method in modeling them is presented. In this way two models of the fable are created, and by comparing them, similarities, differences and complexity of converting to the model of two natural human languages were observed.
9.K. Darabkh, I. Jafar, R. Al-Zubi, M. Hawa (The University of Jordan, Amman, Jordan)
An improved Image Least Significant Bit Replacement Method 
With the development of internet technologies and communication services, message transmissions over the internet still have to face all kinds of security problems. Hence, how to protect secret messages during transmission becomes a challenging issue for most of the researchers. It is worth mentioning that many applications in computer science and other related fields rely on steganography and watermarking techniques to ensure information safety during communication. In this paper, we propose a new steganographic method to embed the secret data inside a cover image based on least-significant-bit (LSB) replacement method. The embedding process predominantly concentrates on distributing the secret message inside one share of a color image to appear like a 3D geometric shape. The dimensions of the geometric shape are variable pursuant to the size of secret message. Data distribution process makes our method to be of a great interest as of being so difficult for the hackers or intruders to reconstruct the shape from stegoimages, thereby the security is improved. Furthermore, we compare the performance of our approach with two other relevant approaches in terms of peak signal-to-noise ratio (PSNR). The contribution of our approach was immensely impressive.
10.K. Darabkh, R. Al-Zubi, M. Jaludi (The University of Jordan, Amman, Jordan)
New Recognition Methods for Human Iris Patterns 
The use of user identification technology has become increasingly in demand in today’s society. Biometrics have been capitalized upon for this purpose. The human iris in particular is one of the most unique and intriguing biometrics available to use in the identification of an individual. The process of recognizing a human iris is split into four major steps. These steps are segmentation, normalization, feature extraction, and matching. In this paper, two new methods are introduced to implement feature extraction. Both methods use a sliding-window technique and different mathematical operations on the pixels to produce feature vectors. Experimental results of the methods produced relatively small feature vectors of size 5x120 and 5x130. A small feature vector is valuable as it contributes to the efficiency and speed of the overall recognition system. In addition, a step was included in both methods to minimize the effect of varying light intensity. This reduces the time needed to acquire an image with suitable lighting, which in turn contributes to the speed of the system as well. Analysis of our methods was done by considering various performance metrics such as the False Acceptance Rate (FAR), False Rejection Rate (FRR), and Recognition Rate (RR). Both proposed methods achieved a recognition rate of about 98.3264%.
Friday, 5/30/2014 9:30 AM - 1:00 PM,
Nava 2, Hotel Admiral, Opatija
Papers 
1.L. Fürst, A. Leonardis (University of Ljubljana, Ljubljana, Slovenia)
Feature Selection for Object Detection: The Best Group vs. the Group of Best 
The problem of visual object detection, the goal of which is to predict the locations and sizes of all objects of a given visual category (e.g., cars) in a given set of images, is often based on a possibly large set of local features, only a few of which might actually be useful for the given detection setup. Feature selection is concerned with finding a ‘useful’ subset of features. In this paper, we compare two approaches to feature selection in a visual object detection setup. One of them selects features based on their individual utility scores alone, regardless of possible interdependence with other features. The other approach employs the AdaBoost framework and hence implicitly deals with interdependence. Using two feature extraction methods and several image datasets, we experimentally confirm the significance of feature interdependence: features that perform well individually do not necessarily perform well as a group.
2.M. Ivašić-Kos, M. Pobar, L. Mikec (Sveučilište u Rijeci, Rijeka, Croatia)
Movie Posters Classification into Genres Based on Low-level Features  
A person can quickly grasp the genre (drama, comedy, cartoons, etc.) from a movie poster, regardless of visual clutter and the level of details. Bearing this in mind, it can be assumed that simple properties of a movie poster should play a significant role in automated detection of movie genres. Therefore, low-level features based on colors and edges are extracted from poster images and used for poster classification into genres. In this paper, poster classification is modeled as a multi-label classification task, where a single movie may belong to more than one class (genre). To simplify and solve the multi-label problem, two methods for multi-label data transformation are described and evaluated given the classification results obtained by distance ranking, Naïve Bayes and RAKEL. Experiments are conducted on a set of 1500 posters with 6 movie genres. Results provide insights into the properties of the discussed algorithms and features.
3.I. Harbaš, M. Subašić (Fakultet elektrotehnike i računarstva, Zagreb, Croatia)
Roadside vegetation detection using features from the visible spectrum 
Vegetation detection in images is a common procedure in remote sensing and is usually applied to satellite and aerial images. Recently it has been applied to images recorded from within ground vehicles for autonomous navigation in outdoor environments. In this paper we present a method for roadside vegetation detection intended for traffic safety and infrastructure maintenance. While many published methods for vegetation detection are using Near Infrared images which are particularly suitable for vegetation detection, our method uses image features from the visible spectrum allowing the use of common on-board color cameras. Our feature set consists of color features and texture features. One of our specific goals was to identify a useful texture feature set for the problem of vegetation detection. Based on the feature set, the detection is implemented using a Support Vector Machine algorithm. For training and testing purposes we recorded our own image database consisting of different images containing roadside vegetation in various conditions. We are presenting promising experimental results and a discussion of specific problems experienced or expected in real-world application of the method.
4.A. Radovan (Veleučilište Velika Gorica, Velika Gorica, Croatia), Ž. Ban (Fakultet elektrotehnike i računarstva, Zagreb, Croatia)
Predictions of Cloud Movements and the Sun Cover Duration 
In order to optimize the microgrid system for direct conversion of solar into electrical energy, it is necessary to determine the exact moment when the clouds block the Sun and create a shadow on the solar panels and how long will it last. The most important parameters in tracking the cloud movement are based on the detection of cloud edges and determine angular velocity vector which describes the speed and direction of cloud movements. A very common occurrences during cloud movement are the change of their shape and the change of direction and speed. To compensate that unpredictable behaviour, the position of every cloud with significant area is approximated with one single point – centroid. This paper analyzes cloud movements using camera with resolution 1920x1080 pixels and refresh rate of 25 frames per second. For the detection of clouds the HSV (Hue-Saturation-Value) color model is used combined with mathematical morphological operations of erosion and dilation. Video and image processing was performed by application developed in Java programming language using the JavaCV library that offers the functionalities of computer vision.
5.M. Simjanoska, M. Gusev, A. Madevska Bogdanova (Ss. Cyril and Methodius University, Faculty of Computer Science and Engineering, Skopje, Macedonia)
Intelligent Modelling for Predicting Students' Final Grades 
The main objective of this paper is producing an intelligent virtual teacher who will be able to predict the students' final grades at the end of the semester. Our approach is based on continual observation of the student's activities on the particular course during the semester. In order to achieve realistic modelling of the students' devotion to the given lectures and also the degree of how much the student has learned from the given lecture, we take into account both the e-Learning and the e-Assessment results through the semester. In our previous work we did an intelligent students' Profiling to classify the students into a pass, or, fail category. In this paper we go deeper into the problem, achieving more precise modelling according to which we will be able to determine the student's most likely final grade, using multi classification methodology. The advantage of our model is in its ability to take into account all the assessments during the semester, not relying only on the results from the last student's assessment. It can be a good indicator whether the teacher needs to perform additional testing of the student's knowledge in order to derive an overall conclusion on the most appropriate grade.
6.B. Trstenjak (Medimurje University of Applied Sciences Cakovec, Cakovec, Croatia), D. Đonko (Faculty of Electrical Engineering/Dept. of Computer Science, Sarajevo, Bosnia and Herzegovina)
Determining the Impact of Demographic Features in Predicting Student Success in Croatia 
Predicting the success of students is a topic which has been studied for a long time in the different scientific fields. Evaluation of importance the features used in the prediction and their subsequent selection is immensely important step in the process of classification and data mining. This paper presents a study on the importance of student demographic features in the process of predicting. The study and performed analyzes were used demographic data collected from students Information System for Higher Education (ISVU). For determining the importance of demographic features in the study has used the following methods: Information Gain (IG), Gain Ratio (GR), Sequential Backward Selection (SBS), Sequential Forward Selection (SFS) and SVM classifier. The results show the features rank, their importance weight in the prediction and comparison of the results towards the use of different methods. Used two classification algorithm for evaluating the impact of ranking features to the quality of prediction: Naive Bayes i Support Vector Machine (SVM). Final results provide guidelines for the development of a new prediction model.
7.J. Tomljanović (Polytechnic of Rijeka, Rijeka, Croatia), M. Pavlić, M. Ašenbrener Katić (Department of Informatics, University of Rijeka, , Rijeka, Croatia)
Intelligent Question – Answering Systems: Review of Research 
With the development of information technology the needs for automated question-answering systems is becoming increasingly important. Question-answering systems are still under development and experimentation . This paper is an overview of the research area that deals with question-answering systems, explaines the concept of question-answering systems and point out the problems that occur during their development. Briefly, it describes the development of the systems to the present current approaches. It also refers to a complex assessment techniques that are necessary when designing such systems. The system descibed is a real system an so are the test results of the same.
8.Z. Balaž, K. Meštrović, P. Jandrlić, M. Čaržavec (Tehničko veleučilište Zagreb, Zagreb, Croatia)
Inteligentni sustav upravljanja tunelskom rasvjetom 
Moguća racionalizacija svakog projekta rasvjete tunela počiva na njegovim utvrđenim eksploatacijskim parametrima. Obradom parametara rasvjete s pozicije ekspertnih sustava, naročito dugačkih tunela, ukazuje da su moguće značajne uštede kako u potrošnji električne energije tako i u održavanju. Raspoložive baze podataka i baze znanja, te manipulacija znanjem tijekom višegodišnje eksploatacije tunelskih rasvjeta rezultirale su kreacijom inteligentnog sustava za upravljanje tunelskom rasvjetom na način drugačiji od uniformno prihvaćenog i oslonjenog na legislative europskih zemalja. Tunelska je rasvjeta kao sustav sačinjena od više međusobno povezanih i zavisnih dijelova rasvjete. Ona u prvom redu pridonosi sigurnosti prometa kroz tunel i stoga je bitno voditi računa o problematici prilagodbe i akceptiranja svih relevantnih parametara luminancije. Inteligentni je sustav zadužen da vodi računa o vizualnoj prilagodbi ali uzimajući u obzir prosječnu luminanciju okoline, samog tunela, predportalnih građevina i prometnice, kao i dnevnog svjetla, neba koje je podložno promjenama kako atmosferskih prilika tako i godišnjih doba. Za tako velik broj utjecajnih parametara razrađene su aktuatorske i perceptorske mogućnosti oslonjene na nove tehnologije opreme, s kojom su provedena ispitivanja. Dobiveni rezultati sintetizirani inteligentnim sustavom su potvrdili opravdanost u novom pristupu projektiranju i izvedbi upravljanja tunelskom rasvjetom.
9.Z. Balaž, K. Meštrović, G. Bijelić (Tehničko veleučilište Zagreb, Zagreb, Croatia)
Inteligentni sustav održavanja aerodromskog sustava svjetlosne signalizacije 
Aerodromski sustavi svjetlosne signalizacije su konfigurirani serijskim krugovima s konstantnim izvorom struje. Važan parametar koji ukazuje na funkcionalnost takvog strujnog kruga je otpor izolacije primarnih kabela serijskih strujnih krugova napajanih iz regulatora konstantne struje. Regulator je opremljen modulima koji nadziru i upravljaju rasvjetom. Raspoloživi parametri se nadziru i arhiviraju u računalo službe održavanja. Kreacijom baza znanja i baza podataka te manipulacijom znanjem postiže se efikasnije održavanje. Inteligentni sustav pomaže u procedurama preventivnog i prediktivnog održavanja i služi za učenje. Stečeno znanje ukazuje na moguće racionalizacije i na unapređenja prilikom revitalizacije ili zamjene postojećih sustava. Razvijenim inteligentnim sustavom koji se koristi u programu edukacije za stjecanje licence aerodromskih specijalista na održavanju sustava svjetlosne signalizacije, dobiveni su kokretni podaci o karakteristikama i efikasnosti pojedinih podsustava. Između ostalih to je sustav za nadzor stanja izolacije serijskih strujnih krugova koji pomaže u podizanju kvalitete izolacije. Kontinuiranim praćenjem i stvaranjem baza podataka o stanju izolacije složeni su algoritni za predviđanje mogućih kvarova a samim tim i za racionalnije održavanje uz visoku pogonsku pouzdanost.
10.D. Vukelić (Gradsko kazalište lutaka, Rijeka, Croatia), M. Franušić, I. Grakalić (Veleučilište u Rijeci, Rijeka, Croatia)
Mobilni uređaji u funkciji racionalizacije upravljanja udaljenim računalima emuliranjem računalnog miša 
S obzirom na razvijenost mobilnih operacijskih sustava i podršku sklopovlju za povezivanje u bežične mreže, pri razvoju aplikacija za mobilne uređaje može se iskorištavati sve aspekte računalnih mreža. Osim mobilnog prijenosa podataka za spajanje na internet mobilnom mrežom, uz pomoć integriranih mrežnih kartica mobilni uređaji spajaju se i u bežične lokalne mreže. Povezivanjem u lokalnu mrežu ostvaruje se komunikacija s ostalim uređajima spojenim u istu mrežu, a time se otvara cijeli niz mogućnosti korištenja tih uređaja putem mobilnog uređaja. Cilj ovog rada je iskoristiti mobilni uređaj u svrhu upravljanja udaljenim računalom uz pomoć pokreta mobilnog uređaja detektiranog ugrađenim akcelerometrom. U radu se definira koncept upravljanja spajanjem mobilnog uređaja i udaljenog računala u bežičnu lokalnu mrežu. Spajanje se odvija u klijent – server arhitekturi putem protokola UDP. Radom je pokazana mogućnost upravljanja emuliranjem rada računalnog miša i dan je pregled mogućih primjena. Moguće primjene sustava temelje se na sve većoj popularizaciji „pametnih“ mobilnih uređaja opremljenih svim potrebnim elementima za rad ovog sustava. Testirana funkcionalnost na više platformi dokaz je valjanosti koncepta i iskoristivosti.
11.D. Lozić, A. Šimec (Tehničko veleučilište u Zagrebu, Zagreb, Croatia)
Pametna komunikacija na Internetu preko REST protokola 
REST protokol je akronim za Representational state transfer protokol koji omogućuje komunikaciju na Internetu pomoću svih najpoznatijih programskih jezika kao što su C/C++, C#, Jave, Ruby, PHP i sličnih kao i preko najpoznatijih operacijskih sustava kao što su Microsoft Windows, Mac OS, Linux itd. Zahvaljujući odličnom radu nad velikim brojem klijenata (engl. scalability), mogućnosti primjene ovog protokola na mnoge programerske probleme kao što su dohvaćanje podataka na serveru, dodavanje, promjena ili brisanje postojećih podatka te neovisnosti o tehnologiji, razlog je zašto baš ovaj protokol koriste mnoge velike tvrtke kao što su Google (GoogleMaps, GoogleSearch), Microsoft (Azure, Shrepoint) za komunikaciju između tehnologija na Internetu. Mogućnosti biranja tehnologije odgovora (json, xml, HTML itd.), mogućnost slanja filtera podataka direktno serveru i kreiranje odgovora u ovisnosti o poslanim HTTP metodama je ono što čini ovaj protokol naprednijim i boljim od bilo koje druge alternative.

Basic information:
Chairs:

Slobodan Ribarić (Croatia), Andrea Budin (Croatia)

International Program Committee Chairman:

Petar Biljanović (Croatia)

International Program Committee:

Alberto Abello Gamazo (Spain), Slavko Amon (Slovenia), Vesna Anđelić (Croatia), Michael E. Auer (Austria), Mirta Baranović (Croatia), Ladjel Bellatreche (France), Eugen Brenner (Austria), Andrea Budin (Croatia), Željko Butković (Croatia), Željka Car (Croatia), Matjaž Colnarič (Slovenia), Alfredo Cuzzocrea (Italy), Marina Čičin-Šain (Croatia), Marko Delimar (Croatia), Todd Eavis (Canada), Maurizio Ferrari (Italy), Bekim Fetaji (Macedonia), Tihana Galinac Grbac (Croatia), Liljana Gavrilovska (Macedonia), Matteo Golfarelli (Italy), Stjepan Golubić (Croatia), Francesco Gregoretti (Italy), Stjepan Groš (Croatia), Niko Guid (Slovenia), Yike Guo (United Kingdom), Jaak Henno (Estonia), Ladislav Hluchy (Slovakia), Vlasta Hudek (Croatia), Željko Hutinski (Croatia), Mile Ivanda (Croatia), Hannu Jaakkola (Finland), Leonardo Jelenković (Croatia), Dragan Jevtić (Croatia), Robert Jones (Switzerland), Peter Kacsuk (Hungary), Aneta Karaivanova (Bulgaria), Dragan Knežević (Croatia), Mladen Mauher (Croatia), Igor Mekjavic (Slovenia), Branko Mikac (Croatia), Veljko Milutinović (Serbia), Alexandru-Ioan Mincu (Slovenia), Vladimir Mrvoš (Croatia), Jadranko F. Novak (Croatia), Jesus Pardillo (Spain), Nikola Pavešić (Slovenia), Vladimir Peršić (Croatia), Goran Radić (Croatia), Slobodan Ribarić (Croatia), Janez Rozman (Slovenia), Karolj Skala (Croatia), Ivanka Sluganović (Croatia), Vlado Sruk (Croatia), Uroš Stanič (Slovenia), Ninoslav Stojadinović (Serbia), Jadranka Šunde (Australia), Aleksandar Szabo (Croatia), Laszlo Szirmay-Kalos (Hungary), Davor Šarić (Croatia), Dina Šimunić (Croatia), Zoran Šimunić (Croatia), Dejan Škvorc (Croatia), Antonio Teixeira (Portugal), Edvard Tijan (Croatia), A Min Tjoa (Austria), Roman Trobec (Slovenia), Ivana Turčić Prstačić (Croatia), Sergio Uran (Croatia), Tibor Vámos (Hungary), Mladen Varga (Croatia), Marijana Vidas-Bubanja (Serbia), Boris Vrdoljak (Croatia), Robert Wrembel (Poland), Baldomir Zajc (Slovenia), Damjan Zazula (Slovenia)

Registration / Fees:
REGISTRATION / FEES
Price in EUR
Before May 12, 2014
After May 12, 2014
Members of MIPRO and IEEE
180
200
Students (undergraduate and graduate), 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 3548
E-mail: andrea.budin@ericsson.com

Opatija - 170 years of tourism:

Opatija – the cradle of European and Croatian tourism, a favourite destination of the aristocracy, film and music stars, artists, writers and visitors from all over the world, who come here every year to enjoy the charm of this Adriatic town – this year celebrates its 170th anniversary as a tourist resort.

This is a tradition that provides certain obligations, but is also a guarantee of quality. The reputation of a top destination that stretches back seventeen decades is today reflected in the wide range of facilities and services on offer that all together make Opatija an attractive destination for all seasons.

Opatija owes its unique image to its ideal location on the spot where the wooded slopes of Mount Učka descend all the way down to the coast, providing perfect shade along the thirteen-kilometre-long Lungomare seafront promenade. Just as the Opatija area is a meeting point of the sea and the mountain, its visual impression is a blend of different styles, as this is a melting pot where magnificent Central European elegance, playful Mediterranean charm and the historically-rich medieval architecture of the small towns in the hinterland come together. 

In addition to the architecture that leaves a strong impression on every visitor, especially when the town is viewed from the sea, and its lush parks and gardens that have been Opatija's trademarks since its beginnings as a tourist resort, Opatija also has hotels and restaurants whose quality ranks alongside that of any other European destination. Opatija's gastronomic offer is based on a Mediterranean cuisine rich in fresh fish and seafood and locally grown seasonal ingredients, while the traditional recipes of this region reveal a wealth of flavours and can be sampled in the area's numerous taverns. 

Opatija entered the European stage in the mid-19th century as a health resort for the European nobility, and health tourism has remained one of the main segments of the town's tourism offer right up to the present day. However, top medical experts and a wide range of spa & wellness services are just one of the reasons for visiting this town located at the top of Kvarner Bay. Also known as "the town of festivals", Opatija boasts a number of events throughout the year. The theatrical performances and concerts that take place at the magnificent Open Air Theatre are particularly impressive.


For more details please look at www.opatija.hr/ and www.opatija-tourism.hr/.

Download
 
News about event
Currently there are no news
 
Patrons - random
Tehnički fakultet RijekaFOI VaraždinIRB ZagrebHAKOMKončar Elektroindustrija Zagreb