Wyniki wyszukiwania - BazTech

1

Efficient line detection method based on 2D convolution filter

Kowalski Paweł, Tojza Piotr

Informatyka, Automatyka, Pomiary w Gospodarce i Ochronie Środowiska

|

2021

|

T. 11, nr 4

22--27

EN

The article proposes an efficient line detection method using a 2D convolution filter. The proposed method was compared with the Hough transform, the most popular method of straight lines detection. The developed method is suitable for local detection of straight lines with a slope from -45˚ to 45˚. Also, it can be used for curve detection which shape is approximated with the short straight sections. The new method is characterized by a constant computational cost regardless of the number of set pixels. The convolution is performed using the logical conjunction and sum operations. Moreover, design of the developed filter and the method of filtration allows for parallelization. Due to constant computation cost, the new method is suitable for implementation in the hardware structure of real-time image processing systems.

PL

W artykule zaproponowano efektywną metodę wykrywania prostych z wykorzystaniem dwuwymiarowego filtru konwolucyjnego. Zaproponowana metoda została porównana z transformatą Hough, najpopularniejszą metodą wykrywania linii prostych. Opracowana metoda pozwala na wykrywanie linii prostych o nachyleniu od -45˚ to 45˚. Może również zostać wykorzystana do wykrywania krzywych, których kształt jest aproksymowany za pomocą krótkich prostych odcinków. Zaproponowana metoda charakteryzuje się stałym kosztem obliczeniowym, niezależnym od liczby pikseli. Splot wykonywany jest z wykorzystaniem logicznej koniunkcji oraz sumowania. Ponadto konstrukcja opracowanego filtru oraz zastosowana metoda filtracji pozwala na zrównoleglenie. Ze względu na stały koszt obliczeniowy, zaproponowana metoda nadaje się do implementacji w strukturze sprzętowej systemów przetwarzania obrazu w czasie rzeczywistym.

2

Initial analysis of the tracking performance of the GOOSE GNSS Software-Defined Receiver

Galas Roman, Kunzi Florian, Adolfs Marjolijn, Overbeck Matthias, Felber Wolfgang

Zeszyty Naukowe Akademii Morskiej w Szczecinie

|

2020

|

nr 63 (135)

113--120

EN

The GOOSE (GNSS Receiver with open software interface) Software-Defined Receiver has been developed at the Fraunhofer Institute for Integrated Circuits (IIS) in Nürnberg, Germany. The main motivation for the development of this platform was to control the receiver at all stages, from digital signal processing to the PVT domain, and to enable controlled feedback to the hardware. Besides having access to all raw data including correlation values, the GOOSE receiver also enables for example tight- or ultra-tight integration with an inertial navigation system or other dead reckoning systems, as these kinds of architectures require access to the acquisition and tracking loops. In this paper, the tracking performance of the GOOSE platform was evaluated and compared to a reference receiver (Septentrio PolaRx5S). Several long data sessions were recorded on a “zero baseline” in which both receivers used the same precise geodetic antenna that was also developed at Fraunhofer IIS. The measurements were performed in a harsh environment (obstructions, multipath, possible interferences), as well as on a site with an unobstructed sky view. Quality and performance analyses were performed using raw measurements (in the domain of primary observables) of three civil GPS signals: L1CA, L2CM, and L5. The data were processed using the “zeroEdit” module of the TUB-NavSolutions academic software for education and research. The quality of the raw observables and tracking performance were described by the following parameters: number of cycle slips detected, number of un-correctable cycle slips, number of loss of locks of the signals, number of single epoch data gaps, and the length of carrier phase arcs. The presentation is illustrated with some numerical examples.

3

Implementing the fade-in audio effect for real-time computing

Lupşa-Tătaru Lucian

Applied Computer Science

|

2019

|

Vol. 15, no 2

5--18

EN

Audio fading is performed in order to smoothly modify over time the level of an audio signal. In particular, the fade-in audio effect designates a gradually increase in the audio volume, starting from silence. In practice, audio fading is mostly carried out within audio editors i.e. in off-line mode by employing various transcendental functions to enforce the fade profile. Taking into account the increasing demand for interactive media services requiring real-time audio processing, the present approach advances an effective method of constructing the audio fade-in shape with a view to real-time computing. The paper encompasses plain and straightforward implementations in pure JavaScript, prepared precisely to validate the method of audio volume processing proposed here.

4

Customizing audio fades with a view to real-time processing

Lupşa-Tătaru Lucian

Applied Computer Science

|

2019

|

Vol. 15, no 4

16--26

EN

To a large extent, an audio fade is distinctly acknowledged as a strict increase (fade-up) or decrease (fade-down) of the volume of an audio content. In this broad context, the widely used fade-in and fade-out sound effects, applied to receive smooth transitions from and down to silence, respectively, appear to be restrictive. Taking into account the increasing demand for multimedia techniques adapted for real-time computing, the present investigation advances straightforward procedures intended for customizing the audio fade-up and fade-down profiles, having at hand well-proven techniques of shaping the fade-in and fade-out audio effects, suitable for fast computing.

5

My City Dashboard: Real-time Data Processing Platform for Smart Cities

Usurelu C.-C., Pop C.

Journal of Telecommunications and Information Technology

|

2017

|

nr 1

89--100

EN

In recent years, with the increasing popularity of IoT, the rapid growth of smartphone usage enabled by the increase adoption of Internet services and the continuously decreasing costs of these devices and services has led to a huge increase in the volume of data that governments can use in the context of smart city initiatives. The need for analytics is becoming a requirement for smart city projects such as city dashboards to provide citizens with an easy to understand overview of the city. As such, data should be analyzed, reduced and presented in such a way that citizens can easily understand various aspects of the city and use this information to increase quality of life. In this paper, we firstly present the context and the start of the design and implementation of proposed solution for real-time data processing in smart cities, mainly an analytics processing pipeline and a dashboard prototype for this solution, named My City Dashboard. We focus on high scalability and modularity of this platform.

6

Development of an autonomous Global Navigation Satellite System ground station and its calibration for monitoring of local ionospheric perturbations

Galas R., Čokrlić M.

Zeszyty Naukowe Akademii Morskiej w Szczecinie

|

2016

|

nr 45 (117)

54--59

EN

A state-of-the-art monitoring global navigation satellite system (GNSS) system has been originally designed and developed for various positioning and atmosphere-sensing purposes by the authors and updated to fulfil the challenging requirements for monitoring of ionospheric perturbations. The paper discusses various scientific and technically challenging issues, such as the requirement for an autonomous operating ground GNSS station and how this can be fulfilled. Basic algorithms for monitoring of local ionospheric perturbations with GNSS receivers are described. The algorithms require that inter-frequency hardware biases be known. Although the satellite transmitter biases can be obtain from the IGS services, the user takes responsibility for the estimation of frequency dependent receiver hardware biases and for the control of their variations. The instrumental signal delays are important for timing applications and GNSS monitoring of the ionosphere and are also required for recovering of the integer carrier-phase ambiguities. The paper presents an algorithm for calibration of inter-frequency biases of global positioning system (GPS) receivers and validates the first set of results.

7

Real-time gas recognition system based on the gas sensor array responses and portable computer Raspberry PI

Ardanowska E., Kalinowski P., Woźniak Ł., Jasiński G., Jasiński P.

Elektronika : konstrukcje, technologie, zastosowania

|

2014

|

Vol. 55, nr 1

16-18

PL

Systemy analizy gazów wykorzystujące czujniki, bądź matryce czujników oraz odpowiednio dobrane algorytmy analizy danych pomiarowych stanowią atrakcyjną alternatywę dla skomplikowanych urządzeń wykorzystywanych w laboratoriach. Umożliwiają rozpoznanie typu gazów występujących w mierzonej przez czujniki mieszaninie lotnych związków oraz stężenia poszczególnych składników tej mieszaniny w środowisku pracy systemu. W artykule zaprezentowano realizację systemu analizy danych pochodzących z matrycy sześciu półprzewodnikowych rezystancyjnych czujników gazu opartego na przenośnym komputerze Raspberry PI. Analiza danych odbywa się w czasie rzeczywistym, przy pomocy napisanego oprogramowania.

EN

The gas-analyzing systems based on the array of partially selective gas sensors and pattern-recognition techniques are potentially fast and low-cost alternative for other devices, like gas-analysers. They give the possibility of recognition the type and the concentration of measured volatile compounds in their working environment. In this work we present the implementation of gas recognition system, in which the signals from an array of six semiconductor gas sensors are processed in real time using the software written for the portable computer Raspberry PI.

8

Skalowalny algorytm estymacji ruchu dla systemów rozproszonych

Konieczny J., Łuczak A.

Pomiary Automatyka Kontrola

|

2009

|

R. 55, nr 8

684-686

PL

W pracy zaprezentowano autorski algorytm umożliwiający realizację procesu estymacji ruchu w czasie rzeczywistym przy wykorzystaniu systemów rozproszonych. Jest to algorytm należący do grupy tzw. szybkich algorytmów estymacji ruchu i umożliwia estymację ruchu z rozdzielczością poniżej jednego okresu próbkowania. Zaproponowany algorytm jest skalowalny i posiada budowę modułową, umożliwia elastyczne dostosowanie do rozmiaru układu oraz udostępnia mechanizm kontroli liczby cykli zegara niezbędnych do estymacji ruchu. W rezultacie możliwe jest dostosowanie częstotliwości zegara do możliwości układu przy zapewnieniu przetwarzania w czasie rzeczywistym.

EN

In this paper a novel scalable algorithm for real-time motion estimation dedicated for distributed systems is presented. The proposed algorithm has modular structure and provides ability to flexibly adjust the global clock-rate required for real-time processing. This is achieved using hierarchical structure of the algorithm, which assumes division of the whole motion estimation process into independent processing stages (Figs. 1, 3, 4) and introducing special mechanism for controlling the allowed number of computation cycles. The algorithm modularity provides additional profits like ability to choose various methods for each processing stage independently and scalability of the circuit structure resulting in more efficient hardware implementation. Flexible clock-rate adjustment enables real-time processing with various types of computational platforms, with special regards to distributed systems consisting of many low-performance units. An example of scalable performance of the algorithm in distributed systems is presented. In order to increase the image resolution in real-time processing the processed image is shared between many processing units (Fig. 5). An exemplary system used for evaluation was created using Digilent "Starter Boards" with Xilinx Spartan-3 XC3S1000 FPGA circuits connected with NOC (Network On Chip) (Fig. 6). In the presented implementation six Spartan-3 circuits were able to estimate motion vectors with half-pel accuracy in real-time for HD resolution (1920x1080) video sequence with 25 frames per second.

9

Image processing system proposed to be applied in a gastroscopic laboratory

Gutek D.

Metrology and Measurement Systems

|

2008

|

Vol. 15, nr 3

381-388

EN

The article presents a way to integrate an endoscope with computer equipment. Software has been proposed to archive and access results of endoscopic tests. The data can be accessed at the hospital area and from outside it. Aside with standard descriptions and conclusions there is also an option to watch pictures and videos recorded during a medical examination. Software for real-time editing of the video material together with the option of applying graphics filters to the watched videos has been proposed for use in endoscopic laboratories.

10

Rekonfigurowalne architektury systemów sprzętowych do przetwarzania i analizy obrazów w pracach AGH

Gorgoń M., Wiatr K., Mikrut Z., Tadeusiewicz R.

Elektronizacja : podzespoły i zastosowania elektroniki

|

2003

|

Vol. 23, nr 7-8

29-33

PL

Publikacja omawia trzy systemy do sprzętowego przetwarzania i analizy obrazów cyfrowych w czasie rzeczywistym, zrealizowane przez autorów w ramach badań prowadzonych na AGH w Krakowie. Wyniki badań, jak i przedstawiona metodologia tworzenia systemów wizyjnych pokazują, że mimo ogromnego postępu technologicznego w elektronice w okresie ostatnich 15 lat, koncepcja budowy dedykowanych systemów wizyjnych, przyjęta w latach osiemdziesiątych a wdrażana i rozwijana w coraz nowszych rozwiązaniach, nie tylko nie traci na wartości, ale coraz pełniej odpowiada zapotrzebowaniu na szybkie i wydajne systemy do przetwarzania sygnałów wizyjnych.

EN

At the paper architectures of three real-time image processing and analysis systems developed at AGH University of Science and Technology are presented. The systems have been the subject of research for more than 15 years. The experience gathered during that period as well as several studies, both theoretical and practical, presented in this paper, has lead the authors to a conclusion that the field of solution concepts for the architectures of dedicated video systems, is much more stable than the rapidly varying (at the pace dictated by the progress of electron technology) realization possibilities of hardware solutions.