Assistance System for Traffic Signs Inventory

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F62156489%3A43110%2F15%3A43909407" target="_blank" >RIV/62156489:43110/15:43909407 - isvavai.cz</a>

Výsledek na webu

<a href="https://ece.pefka.mendelu.cz/sites/default/files/imce/ece_2015_final.pdf" target="_blank" >https://ece.pefka.mendelu.cz/sites/default/files/imce/ece_2015_final.pdf</a>

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Assistance System for Traffic Signs Inventory

Popis výsledku v původním jazyce

We can see arising trend in the automotive industry in last years -autonomous cars that are driven just by on-board computers. During the driverless ride, computer must process a wide set of information gained by GPS locators and computer vision system, including GPS position, driving speed and descriptions of lanes and traffic signs. The traffic signs tracking system must deal with real conditions with data that are frequently obtained in poor light condition, fog and heavy rain or are otherwise disturbed. Completely same problem is solved by mapping companies that are producing geospatial data for different information systems, navigations, etc. Examples of such companies can be TomTom or Mapy.cz. They are frequently using cars equipped with a wide range of measuring instruments including panoramic cameras. These measurements are frequently done during early morning hours when the traffic conditions are acceptable. However, in this time, the sun position is usually not optimal for the photography. Most of the traffic signs and other street objects are heavily underexposed. Hence, it is difficult to find an automatic approach that can identify them reliably. In this article, we focus on methods designed to deal with described conditions. An overview of the state-of-the-art methods is outlined. Further, where it is possible, we outline an implementation of described methods using well-known Open Computer Vision library. Finally, emphasis is placed on methods that can deal with low light conditions, fog or other situations that complicate the detection process.

Název v anglickém jazyce

Assistance System for Traffic Signs Inventory

Popis výsledku anglicky

We can see arising trend in the automotive industry in last years -autonomous cars that are driven just by on-board computers. During the driverless ride, computer must process a wide set of information gained by GPS locators and computer vision system, including GPS position, driving speed and descriptions of lanes and traffic signs. The traffic signs tracking system must deal with real conditions with data that are frequently obtained in poor light condition, fog and heavy rain or are otherwise disturbed. Completely same problem is solved by mapping companies that are producing geospatial data for different information systems, navigations, etc. Examples of such companies can be TomTom or Mapy.cz. They are frequently using cars equipped with a wide range of measuring instruments including panoramic cameras. These measurements are frequently done during early morning hours when the traffic conditions are acceptable. However, in this time, the sun position is usually not optimal for the photography. Most of the traffic signs and other street objects are heavily underexposed. Hence, it is difficult to find an automatic approach that can identify them reliably. In this article, we focus on methods designed to deal with described conditions. An overview of the state-of-the-art methods is outlined. Further, where it is possible, we outline an implementation of described methods using well-known Open Computer Vision library. Finally, emphasis is placed on methods that can deal with low light conditions, fog or other situations that complicate the detection process.

Druh

D - Stať ve sborníku

CEP obor

IN - Informatika

OECD FORD obor

—

Projekt

—

Návaznosti

I - Institucionalni podpora na dlouhodoby koncepcni rozvoj vyzkumne organizace

Rok uplatnění

2015

Kód důvěrnosti údajů

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

Název statě ve sborníku

Enterprise and Competitive Environment: Conference Proceedings

ISBN

978-80-7509-342-4

ISSN

—

e-ISSN

—

Počet stran výsledku

10

Strana od-do

1008-1017

Název nakladatele

Mendelova univerzita v Brně

Místo vydání

Brno

Místo konání akce

Brno

Datum konání akce

5. 3. 2015

Typ akce podle státní příslušnosti

EUR - Evropská akce

Kód UT WoS článku

000380464000109