Radiotekhnika
Publishing house Radiotekhnika

"Publishing house Radiotekhnika":
scientific and technical literature.
Books and journals of publishing houses: IPRZHR, RS-PRESS, SCIENCE-PRESS


Тел.: +7 (495) 625-9241

 

Application of the algorithm for detecting objects using interactively defined color and texture attributes for recognizing traffic signs on the video stream

DOI 10.18127/j20700784–201808–04

Keywords:

A.S. Maksimovskiy – Post-graduate Student, P.G. Demidov Yaroslavl State University
E-mail: maxal9999@mail.ru
A.L. Priorov – Dr.Sc (Eng.), Associate Professor, P.G. Demidov Yaroslavl State University
E-mail: andcat@yandex.ru
Yu.A. Bryukhanov – Dr.Sc (Eng.), Professor, P.G. Demidov Yaroslavl State University
E-mail: bruhanov@uniyar.ac.ru


This work describes a technique based on the analysis of color and textural signs of road signs on the image, which allows them to be selected in the image, but also to recognize the type of the sign. This proposed methodology is implemented in the C++ programming language using the OpenCV library.
The analysis of texture attributes was performed with the help of a trained neural network. The peculiarity of training the neural network using the Voronoi diagram is considered. Also the analysis of the dependence of the probability of correct detection of road signs on illumination in the aid is given, and the technique for reducing this dependence is proposed.

References:
  1. Shapiro L., Stokman Dzh. Komp'yuternoe zrenie / Per. s angl. M.: BINOM. Laboratoriya znanij. 2006.
  2. Maksimovskij A.S. Obnaruzhenie dorozhnyh znakov na videopotoke v nochnoe vremya sutok na osnove teksturnyh i cvetovyh priznakov // Materialy 18–j mezhdunar. konf. «Cifrovaya obrabotka signalov i ee primenenie – DSPA–2016». S. 948–954.
  3. Maksimovskij A.S., Palochkin V.V. Detektirovanie ob'ektov po interaktivno zadavaemym teksturnym i cvetovym priznakam s pomosch'yu obucheniya nejronnoj seti // Mezhdunarodnyj nauchnyj institut «Education». Ezhemesyachnyj zhurnal. 2014. № 5. S. 103–107.
  4. Maksimovskij A.S., Palochkin V.V. Raspoznavanie interaktivno zadannyh dorozhnyh znakov sredi neskol'kih na osnove teksturnyh i cvetovyh priznakov s pomosch'yu nejronnoj seti // Mezhdunarodnyj nauchnyj institut «Education». Ezhemesyachnyj zhurnal. 2015. № 8 (15). Ch. 2. S. 81–86.
  5. Osovskij S. Nejronnye seti dlya obrabotki informacii / Per. s pol'skogo I.D. Rudinskogo. M.: Finansy i statistika. 2002. S. 78–80.
  6. Neural networks [Elektronnyj resurs]. URL: http://www.alglib.net/dataanalysis/neuralnetworks.php (data obrascheniya: 20.10.16).
  7. ALGLIB Manual C# [Elektronnyj resurs]. URL: http://www.alglib.net/translator/man/manual.csharp.html (data obrascheniya: 20.10.16).
  8. Gonsales R., Vuds R., Eddins S. Cifrovaya obrabotka izobrazhenij v srede MATLAB. M.: Tehnosfera. 2006.
  9. Maksimovskij A.S., Palochkin V.V. Issledovanie kachestva raspoznavaniya ob'ektov s zadannymi pol'zovatelem priznakami v zavisimosti ot osveschennosti i rasstoyaniya do kamery // Prospero. 2015. № 8 (20). S. 22–26.

© Издательство «РАДИОТЕХНИКА», 2004-2017            Тел.: (495) 625-9241                   Designed by [SWAP]Studio