Monthly Archives: May 2010

Shot Boundary Detection and Camera Motion Estimation in Video Sequence

Stanislav Panev, Agata Manolova

Abstract: In this paper we propose a combined algorithm for automatic shot boundary detection and camera motion estimation in video sequence. For the shot boundary detection we use the best and most balanced algorithm based on histogram intersection between successive frames and we determine the adaptive threshold for shot boundary detection. The principle of estimating the camera motion is based on the number of macro blocks having motion vectors of non-zero module and the total number of macro-blocks in the frame. Experimental results on real video sequences confirm the effectiveness of the proposed algorithm in terms of speed and accuracy.

Key words: adaptive threshold, camera motion estimation, color histogram intersection, shot detection

Download Article

ECG QRS Identification Using Morphological Filtering and Histogram Analysis

Yuliyan Velchev

Abstract: This article presents an approach for identification of QRS complexes in human electrocardiogram (ECG). This approach incorporates morphological filtering and histogram analysis-based adaptive threshold criterion. The thresholds are found using Otsu’s method, which is performed on detected stationary points of the filtered signal. The proposed algorithm is fast, reliable and doesn’t need any training procedures. Also it is insensitive to possible baseline drift and baseline wandering. The algorithm is evaluated with real world ECG signals taken from own database. The sensitivity of the achieved detection rate is high.

Key words: ECG analysis, morphological filtering, QRS detection

Download Article

Throughput Improvement in MIMO OSTBC OFDM Systems

Miglen Ovtcharov, Pavlina Koleva

Abstract: In this paper we propose a scheme for throughput improvement in Multiple Input Multiple Output (MIMO) Orthogonal Space-Time Block Coded (OSTBC) Orthogonal Frequency Division Multiplexing (OFDM) systems by Narrow Band Interference (NBI) suppression, using Frequency Excision (FE) algorithm. The experiments outline that the OFDM link performance depends strongly on diversity parameters of the MlMO system as well as on the NBI suppression method efficiency.

Keywords: NBI, MIMO, OFDM, OSTBC, Complex Signals.

Download Article

SCP-RPSC Technology – the Perfect Solution of the Broadband MSS Problems

Veselin Demirev

Abstract: A new technology, named SCP-RPSC, was developed recently for consumer broadband interactive satellite communications. It uses entirely new approach, based on random phased antenna arrays and correlation processing in the receiver. The goals of this report are to discuss the possibilities and the advantages of SCP-RPSC technology from effective frequency sharing point of view, to consider possible improvements of the regulatory status of the satellite services, using vehicle mounted SCP-RPSC antennas, as well as to discuss the ways to protect them from space terrorism.

Key words: SCP-RPSC Technology Mobile Satellite Communications

Download Article

Expressiveness Evaluation of XML-based Language for Service Logic Description

Evelina Pencheva, lvaylo Atanasov

Abstract: There exist several XML-based languages for telecommunication service creation which scope covers a limited number of network functions. A new markup language for service logic description is developed that is oriented toward the application programming interfaces of Parlay/ Open Service Architecture and makes use of the whole palette of network functions exposed by the telecom operator. The expressive power of the language is evaluated using formal grammars and tree automata. The language is described as a homogeneous unambiguous regular tree language and its expressiveness is compared to Document Type Definitions (DTD) and XML schemas.

Key words: formal languages, open service access, regular grammars, service creation, tree automata

Download Article

Mammographic Image Segmentation by Thresholding and Fuzzy C-means Clustering

Antoaneta Popova, Vladimir Kanchev, Ivo Draganov

Abstract:The aim of this paper is to present an algorithm for automatic detection and segmentation of malignant lesions in X – Ray mammographic images. We propose an approach of abnormal mass detection that employs morphological operations for the creation of morphological gradient image of mammograms, binarization with an adjusting threshold, extraction of features eccentricity and area of segmented objects and Fuzzy C-Means clustering on them. We use classification rules based on additional features as number of clusters and objects in order to obtain two classes: region of abnormalities and region of healthy breast tissue. Parameters of the clustering and segmented regions are employed for more accurate intensity threshold calculation.

Key words: automatic image segmentation, breast mass detection, digital mammography, fuzzy C-Means (FCM) clustering, malignant lesion.

Download Article

Management of Database for Processing of the Test Control Results for Assessment of the Knowledge

Daniela Minkovska, Jordan Markov

Abstract: The paper presents the management process trough the database management system of a relation database that consist of information gathered from results of test control of assessment of the knowledge of the students. The rules for creating of the model and entity relational diagram, visualizing the associative relations between the objects into database are described. The process of preservation and processing of the data with the help of DBMS- PostgreSQL is presented. The conclusions are drawn for the application and adaptation of the developed DBMS in the study process and for its future implementation through application for visualizing of the results.

Keywords: database, DBMS, tests, knowledge

Download Article

Embedded System for Express Diagnostics of Accumulators and Batteries

Kamelia Raynova, Anastasia Vangelova

Abstract: In today’s portable and autonomous computer systems, for example in hybrid cars with electric propulsion, mobile phones and many other applications, electrochemical power sources – accumulators and batteries, plays a crucial role for their reliable work. Measurement of their parameters usually requires complex and expensive instrumentation and is often impossible to make “in situ”. In this paper is proposed an express method and embedded computer system for measuring the internal resistance of such sources as an essential indicator of their performance – state of charge and life cycle.

Keywords: impedance, EIS, diagnostics, embedded systems, CAN, IEEE 1451 , TEDS, smart transducers

Download Article

Presentation of Scalar and Vector Data for Visualization in Different Programming Environments

Virginiya Dimitrova, Mariana Goranova, Bogdan Shishedjiev

Abstract: The presented article explains the results achieved in the process of creation of a formalized description of scalar and vector data. The goal was to create such a description that will allow the separate tasks concerning the visualization of field’s distribution and behavior to be realized from separate program modules. One important requirement to the modules is the possibility to be implemented using different program languages and different program environments. To satisfy the requirements for broad accessibility. the modules are implemented as Web-services and the suggested formalized scalar and vector data description is intended to support an effective communication between different visualization services. This will allow creating scalable applications that support a comprehensive set of possibilities for visualization of scalar and vector data.

Key words: communication protocols, field distribution, scalar data, vector data, Web-services, XML-document

Download Article

Information Security and e-Government Cloud Computing Applications

Roumen Trifonov

Abstract: Cloud computing, is a new paradigm for obtaining computing resources and services. In a cloud computing model the resources, such as storage and processing power, as well as services, including software platforms and applications, are dynamically accessed through (most often, public) networks. The cloud’s attributes of scale and flexibility are both a friend and a foe from a security point of view. The massive concentrations of resources and data present are more attractive target to attackers, but cloud-based defences can be more robust, scalable and cost-effective.

Key words: Information security, risk assessment, vulnerabilities, cloud computing

Download Article