GPU Accelerated One-pass Algorithm for Computing Minimal Rectangles of Connected Components

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21230%2F11%3A00177654" target="_blank" >RIV/68407700:21230/11:00177654 - isvavai.cz</a>

Výsledek na webu

<a href="http://ieeexplore.ieee.org" target="_blank" >http://ieeexplore.ieee.org</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1109/WACV.2011.5711542" target="_blank" >10.1109/WACV.2011.5711542</a>

Jazyk výsledku

angličtina

Název v původním jazyce

GPU Accelerated One-pass Algorithm for Computing Minimal Rectangles of Connected Components

Popis výsledku v původním jazyce

The connected component labeling is an essential task for detecting moving objects and tracking them in video surveillance application. Since tracking algorithms are designed for real-time applications, efficiencies of the underlying algorithms become critical. In this paper we present a new one-pass algorithm for computing minimal binding rectangles of all the connected components of background foreground segmented video frames (binary data) using GPU accelerator. The given image frame is scanned oncein raster scan mode and the background foreground transition information is stored in a directed graph where each transition is represented by a node. Further we use GPU acceleration to speed up feature extraction from the image to a directed graph fromwhich minimal bounding rectangles will be computed subsequently. Also we compare the performance of GPU acceleration (Tesla C2050) with the performance of multi-core (up 24 cores) general purpose CPU implementation of the algorithm.

Název v anglickém jazyce

GPU Accelerated One-pass Algorithm for Computing Minimal Rectangles of Connected Components

Popis výsledku anglicky

The connected component labeling is an essential task for detecting moving objects and tracking them in video surveillance application. Since tracking algorithms are designed for real-time applications, efficiencies of the underlying algorithms become critical. In this paper we present a new one-pass algorithm for computing minimal binding rectangles of all the connected components of background foreground segmented video frames (binary data) using GPU accelerator. The given image frame is scanned oncein raster scan mode and the background foreground transition information is stored in a directed graph where each transition is represented by a node. Further we use GPU acceleration to speed up feature extraction from the image to a directed graph fromwhich minimal bounding rectangles will be computed subsequently. Also we compare the performance of GPU acceleration (Tesla C2050) with the performance of multi-core (up 24 cores) general purpose CPU implementation of the algorithm.

Druh

D - Stať ve sborníku

CEP obor

JA - Elektronika a optoelektronika, elektrotechnika

OECD FORD obor

—

Projekt

—

Návaznosti

Z - Vyzkumny zamer (s odkazem do CEZ)

Rok uplatnění

2011

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

Applications of Computer Vision (WACV), 2011 IEEE Workshop

ISBN

978-1-4244-9496-5

ISSN

1550-5790

e-ISSN

—

Počet stran výsledku

6

Strana od-do

479-484

Název nakladatele

IEEE Computer Society

Místo vydání

Los Alamitos

Místo konání akce

Kona, Hawaii

Datum konání akce

5. 1. 2011

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—