Ebook: Intelligent Interactive Multimedia Systems and Services

This volume summarizes the works and new research results presented at the Sixth International Conference on Intelligent Interactive Multimedia Systems and Services (KES IIMSS2013) which took place in June 26-28, 2013 Sesimbra, Portugal. KES IIMSS is a well established international annual conference within the KES International series of conferences that focuses on research in the fields of intelligent interactive multimedia systems and services. The aim of the IIMSS conference is to provide an internationally respected forum for scientific research in the technologies and applications of intelligent interactive multimedia systems and services.

IIMSS2013 consisted of keynote talks, oral presentations, invited sessions and workshops, on the applications and theory of intelligent interactive multimedia systems and services and related areas. IIMSS2013 provided excellent opportunities for the presentation of interesting new research results and discussion about them, leading to knowledge transfer and generation of new ideas.

IIMSS2013 received many submissions of high quality. The submitted papers were reviewed by at least two reviewers. Not all submissions could be accommodated for presentation at the conference. Finally, 37 papers were accepted for presentation at the IIMSS2013 and inclusion in its Proceedings. We are very satisfied with the quality of the program and would like to thank the authors for choosing KES-IIMSS as the forum for presentation of their work. Also, we gratefully acknowledge the hard work of KES-IIMSS international program committee members and of the additional reviewers for taking the time to review the submitted papers rigorously and select the best among them for presentation at the conference and inclusion in its proceedings.

We are also grateful to the KES personnel for their wonderful work in maintaining the IIMSS2013 site. Finally, we would like to thank the IOS Press personnel for their wonderful job in producing this Proceedings.

The General and Program Co-Chairs:

George A. Tsihrintzis, University of Piraeus, Greece

Maria Virvou, University of Piraeus, Greece

Toyohide Watanabe, Nagoya University, Japan

Lakhmi C. Jain, University of South Australia, Australia

The Executive Chair:

Robert Howlett, KES International & Bournemouth University, UK

The Local Arrangements Chair:

Ana Rita Campos, UNINOVA, Portugal

Supporting aggregate range queries on remote spatial databases suffers from 1) huge and/or large number of databases, and 2) limited type of access interfaces. This paper proposes Regular Polygon based Search Algorithm (RPSA) to overcome these difficulties. RPSA requests a series of k-NN queries to obtain approximate aggregate range query results. The query point of a subsequent k-NN query is chosen among vertices of a regular polygon inscribed in a before-searched circle. Experimental results on sum range query search show that Precision is over 0.99 for uniformly distributed dataset, over 0.97 for skew-distributed dataset, and over 0.97 for real dataset. Also, Number of Requests (NOR) ranges between 3.1 and 3.9, between 3.4 to 4.3, and between 2.7 and 3.7, respectively.

The technique of video stabilization removes intentional motions from video sequences caused by vibrations and other artifacts during a hand-held shooting. In this paper, the task of digital video stabilization of static scenes is investigated. Fuzzy logic is used for improvement of local and global motion estimations and determines the novelty of such approach. Three strategies were used for local motion estimations. They are based on two types of keypoint detectors and block-matching algorithm. The application of fuzzy logic operators improves the separation results between the unwanted motion and the real motion of rigid objects. The corrective algorithm compensates the unwanted motion into frames; thereby the static scene is aligned. The quality of test videos stabilization was estimated by PSNR (Peak Signal to Noise Ratio) and ITF (Interframe Transformation Fidelity) metrics. Experiments show that ITF average estimations increase on 5-11% relative to original video sequences.

The efficient management in Digital e-Library Warehouse (DLW) causes the wide challenges from modern architecture design to advanced customization and personalization. The paper provides the multi-level content-dependent architecture of DLW. Also EPUB (Electronic PUBlication) format extensions were investigated which prefer good multimedia possibilities based on advanced visualization technologies especially for end users working in groups. The usability of DLW is based on skill acquisition with collaboration of advanced software/hardware technologies. The experimental tests on vulnerabilities permitted to choice a more appropriative decision – Content Management System Bitrix Framework as a middleware into proposed multilevel architecture of DLW.

One of the major concerns of integrated utilization of Web-based rich multimedia systems and mobile computing technologies is a low security level of communications and transfer of confidential rich multimedia files over mobile networks. Virtual Private Network technology proved to be one of the most reliable technologies to provide data protection, confidentiality, integrity, data origin authentication, replay protection and access control in mobile computing. This paper presents the outcomes of research, design and development project on creation of secure communications in Web-based rich multimedia systems and performance analysis of encryption algorithms in cipher-block chaining mode in IPsec-based mobile virtual private networks.

Recent advances in serious games and computer game engines' development as well as successful applications of serious games concepts/mechanics/mechanisms in corporate management, administration, business, marketing, training, learning, education, wellness and ecology initiatives, and healthcare stimulated initial “serious games” (or, “gamification”) research, design and development. The goals of this conceptual paper are to a) describe the current status of gamification, b) provide designed and developed examples of successful applications of serious computer game development technologies to produce quality knowledge and skills in computer science, and c) show that a collaborative work of various specialists in computer science – software engineers, advanced programmers, graphic user interface designers, human-computer interaction experts, specialists in computer graphics, networking, security databases, and artificial intelligence – is necessary to produce quality developers (technologists) of serious games and gamification applications for real-world simulations and education (edutainment).

We propose a retrieval method of article-pages by specifying the structural feature of page layout. Traditionally, keyword-oriented retrieval means have been used to select the appropriate article-pages from document databases or necessary data from many accumulated data resources: for example, the retrieval mechanism in Google is most popular and effective. However, keywords may be not applicable to retrieval means required for images because images cannot be always identified explicitly by particular words. Also, keywords are not useful for many foreigners who cannot make use of Japanese keywords familiarly in finding out Japanese materials, or for children who cannot remember the technical/difficult keywords easily because main words or phrases are not yet learned. On the other hand, our method is different from these traditional methods, and mainly focuses on the layout-based positions of non-text components in article-pages such as figures, tables, photographs, etc. Under such a layout-oriented approach, our research objective is to support partial-match search means based on human spatial cognition ability, so as to complement the traditional keyword-oriented means.

This paper presents a method of using a classification procedure for retrieval of the messages that may be interesting to students. Students and professors are communicating using an online educational environment, which will provide the necessary amount of data needed to create the student model. We aim at creating a tool that will allow students to receive a set of recommended messages that may be interesting to them. The system makes the recommendations to students based on their activity performed on the platform. The tool analyses the student's features, classifies him and according with obtained class the retrieved set of messages is obtained. The tool also acts as a message indexing system by storing messages in correspondence with student labels that are obtained by employing a decision tree classifier. The data used to train the classifier is gathered from the on-line educational environment and contains all the necessary information regarding the students' performed activities on the platform.

The paper describes a method for acquiring and visualizing the Polish Sign Language gestures along with mimic sub code. The software allowing visualization of sign language gestures is one of the modules of the system for the translation of texts written in the Polish language into appropriate messages of the sign language. Proper understanding of the information communicated in the sign language requires the information to be presented in the multipath manner. In addition to the ideographic communication, i.e. gestures of the sign language alone, also lip movement (many deaf people are able to read the movement of the lips) and elements of the mimic sub code expressing emotions and intentions of the sign language users should be taken into account. Additional paths improving quality of the communication are: the text displayed during the gestures visualization and the audio channel through which the speech synthesis is realized (most deaf people are not completely deaf and the sounds they hear support the communication process). In order to obtain the most accurate and the fullest understanding of the content in the sign language all these channels are transmitted simultaneously. The paper describes process of the acquisition of sign language gestures, modeling of face mimicry, representation and storing of the animation data in the database and visualization of this data using 3D avatar.

The massive use of Facebook opens new horizons in the areas of information and communications technology, which induced changes in pedagogy. Towards this direction, we implemented an educational application for learning languages. In order to cover the heterogeneous group of people to which educational applications are targeted and to further emphasize the educational aspect of Facebook, we implemented and present the multivariate clustering for group learning in Facebook. By the incorporation of the multiple parameter based clustering, the problem of grouping students, in order to provide sophisticated user models, is further treated. Multivariate clustering is conducted by the k-means algorithm which takes as input, to initialize the process, multiple students' characteristics.

The paper presents the solution adopted to index the information in a Multimedia Database Management System dedicated to medical domain, being optimized for medium sized databases. The system automatically processes the images inserted into database, extracting characteristics as: color histogram, texture, and color regions. The system contains the needed algorithms for images processing and visual content-based retrieval, thus making the client applications less complex. The information is indexed using a single level B-tree structure for text-based data, and using an R-trees structure for multimedia data. The performances of the system are analyzed from the execution time point of view.

We propose a multi-stage blind signal detector for spread spectrum signals suitable for wireless ad hoc network nodes. The proposed method separately detects all signals from surrounding nodes after RAKE combining using neither training sequences nor signal timing information. We verify the performance of the proposed method through computer simulations.

With the advent of online education services such as iTunes U and SlideShare, students are able to access various presentation files from many universities for study purposes. However, self-learners retrieving such files still need support to decide which files are worth learning because it can be difficult to understand the context of each file. In this paper, we introduce a quick browsing system to help a user effectively decide whether a presentation content is appropriate for his/her specific learning needs. Our method provides context-based word clouds that summarize slide information to help the user visually understand the context of a presentation content. In our method, words important to the “presentation context,” that is, the relevant information on the slides, are first extracted based on components of the presentation content (i.e., intra-slide structure and inter-slide structure). Then the word cloud is generated by weighting the words within the presentation context for each slide. Finally, the word clouds are interactively presented with visual effects that reflect their semantic meanings.

In order to aid improvements in teaching, the full text of digital course materials used for read-through and quizzes on basic statistics were created on Moodle, to be used to gather learning data. Using Moodle, the aggregate learning data about the students was statistically analyzed. When using digital course materials and quizzes in the classroom, a considerable correlation was observed between the ratios that checked digital course materials and quiz scores. Next, reading times and quiz scores are compared, and also the correlation between the number of page views and quiz scores. Therefore, accessing digital course material and quizzes on Moodle was shown to have a good effect on the students' study.

This paper describes a method for problem-posing and proposes a system for assisting to operate the method. It is important for students to pose diverse problems, although posing diverse problems is difficult. In the proposed method, the students pose problems by modifying examples provided by the teacher, and they evaluate posed problems by peer on the basis of the diversity of situations and solution. Evaluations behave as opportunities for the students to consider the diversity of problems. According to the experimental result, the effect of the method is confirmed.

To exhibit quasi-scientific objects in an art museum is not a new phenomenon. However, both experts and visitors repeat the same behavior familiar to each of them. It causes a deadlock situation. In order to avoid this deadlock, I propose that we must approach such exhibits from the so-called ‘semantic view’ of scientific theories. First, I point out that the problem from M.C. Beardsley's traditional aesthetics comes from his ‘classical’ view of scientific theories. Second, I introduce the ‘semantic’ view of scientific theories and its key concept of model, a concept that solves the problem of the classical view. Then, I show that, if visitors apply the semantic view to quasi-scientific exhibits, visitors could benefit and would be liberated from the deadlock. At the same time, I argue against objections to the semantic view. I also discuss my proposal.

This paper examined the generic texts in scientific and technical communication since the Fukushima Daiichi nuclear disaster from the perspective of its social implications. The analysis revealed that persuasive rhetoric is one form of paternalism that occurs in communication and suggests the importance to switch from the communication to persuade the citizens to the communication to aids the citizen-based decision-making process.

We propose an intelligent interface for the mobile software agents that we have developed. The interface should have two roles. One is to visualize the mobile software agents using augmented reality and the other is to give a human user the means to control the mobile software agents by gesture using a motion capture camera. Through the interface we human beings can intuitively grasp the activities of the mobile agents, i.e. through augmented reality. In order to provide proactive inputs from the user, we utilize the Kinect motion capture cameras to capture the human users' will. The Kinect motion capture camera is mounted on a mobile robot that is near the human user. The robot acts as a mediator that recognizes the human user's will and convey it to mobile software agents that control the mobile multiple robots. A mobile software agent is searching a target, and uses a mobile robot to seize the target when it finds the target. When the user points at the target or a mobile robot, the monitoring software captures the will of the user and conveys instructions to the mobile agent based on the information from the Kinect. The mobile agent migrates from one robot to another to look for the searching mobile agents and hands the instruction to which robot it should move. The agent migration is represented by an image's moving to the robot that was pointed. This paper reports the intelligent user interface that provides the interaction between the human user and the mobile agents as the first step toward the complete intelligent human computer interface. We demonstrate the usefulness through preliminary experiments.

In this paper we examine books published for the general public after the Fukushima Daiichi nuclear disaster to clarify the characteristics, problems, and new trends in the scientific communication related to the health effects of radiation..

An important issue in education systems is the ability to determine the characteristics of learners and then provide intelligent and informed guidance in response. The authors of this paper have a long-term research goal to provide language learners with the ability to determine and improving their weaknesses. However, to achieve this goal a sizable amount of manually classified data is required. The task is both time consuming and labor intensive. In this paper a system was built to help intelligently classify the errors in an English learner's writings into categories (Kroll 1990, Weltig 2004). Using a randomly selected manually classified sample as training data, it was determined that there is a positive correlation between the number of samples for each error category and the effectiveness of the model created by applying SVM machine learning to the writings of language learners on the Lang-8 website. It is intended that the classification results will be used to accelerate the manually process classification and increase the amount of training data available for use.

In Japan, agricultural activation poses a big problem and the increase in new-entrants-to-agriculture persons is expected. This paper analyzes questionnaire surveys of those who want to start working in agriculture to figure out the difficulties and anxiety they have. The questionnaires consist of the categorical data, such as status of their willingness to be a farmer, their age and sex and the free texts that are written by person who answered several sets of questions. Categories and free text question items are considered as viewpoints in this paper. Two viewpoints are selected to construct a cross tabulation of a search result. The number of the questionnaires that matches the two conditions is displayed in the cell of the cross table. All possible cross tables can be shown as a result. A specified cross table is shown if the user determines a pair of viewpoints. This paper reports case studies which are hard to achieve by simple keyword search.

In our previous work, we developed AirTransNote, a student note-sharing system that facilitates collaborative and interactive learning during regular lectures in conventional classrooms. Because taking notes on paper is a regular activity, our system does not impose an extra burden on students who share notes. However, in order to improve the effectiveness of sharing notes on peer learning, students need to improve the presentation of their handwritten notes, because carefully written notes are more accessible for other students and reduce irrelevant, careless mistakes. To facilitate improvements in note-taking, we applied a set of metrics that determine how carefully the notes were written, and propose a system that provides feedback to students about how carefully they are writing notes.

With the increasing number of bicycle riders, there is an increased risk of accidents between bicycles and automobiles. Some accidents are caused by speeding bicycles and the lack of communication between the cyclist and car drivers. To close the communication gap, we propose a method to detect speeding behaviors using a conventional smartphone. Using the smartphone, the cyclist will be able to obtain additional notification with minimum inconvenience and cost. We investigated the fundamental characteristics of accelerometer data with different mounting positions and shock absorber conditions. We found that the rear side position was the best for detecting speeding behaviors.

The article addresses the problem of intelligent monitoring of facility management data flows collected from heterogeneous sources, including low-level data of sensors and probes, geographical indicators, scheduling and personal identification systems. We suggest the framework of analytical model, based on deriving descriptors which could sentinel the level of thermal comfort of working environments. The model aims to facilitate process of extracting essential characteristics of facility management for detecting dependencies and observing anomalies. The performance of the model was tested by experimental analysis of facility management of the university campus, designed for exploring how various environment variables affect temperature in the lecture rooms, equipped by the air conditioning devices.

This paper proposes a method to improve factor models with text information based on the factor model construction process. We analyze the model using the KJ method. However, because this model often has many unidentifiable latent variables due to the lack of observed variables, we need to filter and change the model. To decrease the number of unidentifiable latent variables, we use principal components analysis for keywords in text data and the principal component scores are allocated to the unidentifiable latent variables. Further, we could confirm that GFI (goodness-of-fit index) and AGFI (adjusted goodness-of-fit index) are improved by using the factor model for an online-game analysis.