2018 7th Mediterranean Conference on Embedded Computing (MECO)

Usage of fully convolutional network with clustering for traffic light detection

Yudin D. 1
Slavioglo Dmitry 1
1
 
[Department of Technical Cybernetics, Belgorod State Technological University named after V.G. Shukhov (BSTU n.a. V.G. Shukhov), Belgorod, Russian Federation]
Publication typeProceedings Article
Publication date2018-06-01
Quartile SCImago
Quartile WOS
Impact factor
ISSN
Abstract
In this paper we consider a traffic light detector constructed on the basis of a fully convolutional neural network for segmenting traffic lights on image and subsequent clustering, which allows us to obtain bounding boxes for traffic lights. The proposed approach is compared with one of the most effective object detectors — Single Shot Multibox detector (SSD). We implemented algorithms for objects detection on an embedded system based on the NVidia Jetson TX2 platform. Traffic light detection recall for the proposed approach is better than SSD and higher than 0.9 on both the testing and training samples from relatively small data set (500 training images and 107 testing images). Time of traffic lights detection on one frame is about 50 ms. The results of the traffic light detection prove the possibility of applying the approach based on fully convolutional neural network with clustering for embedded autonomous vehicle control systems and driver assistance systems.

Citations by journals

1
Optical Memory and Neural Networks (Information Optics)
Optical Memory and Neural Networks (Information Optics), 1, 9.09%
Optical Memory and Neural Networks (Information Optics)
1 publication, 9.09%
Buildings
Buildings, 1, 9.09%
Buildings
1 publication, 9.09%
IEEE Access
IEEE Access, 1, 9.09%
IEEE Access
1 publication, 9.09%
Smart Innovation, Systems and Technologies
Smart Innovation, Systems and Technologies, 1, 9.09%
Smart Innovation, Systems and Technologies
1 publication, 9.09%
Lecture Notes in Computer Science
Lecture Notes in Computer Science, 1, 9.09%
Lecture Notes in Computer Science
1 publication, 9.09%
AIP Conference Proceedings
AIP Conference Proceedings, 1, 9.09%
AIP Conference Proceedings
1 publication, 9.09%
1

Citations by publishers

1
2
IEEE
IEEE, 2, 18.18%
IEEE
2 publications, 18.18%
Springer Nature
Springer Nature, 2, 18.18%
Springer Nature
2 publications, 18.18%
Pleiades Publishing
Pleiades Publishing, 1, 9.09%
Pleiades Publishing
1 publication, 9.09%
Multidisciplinary Digital Publishing Institute (MDPI)
Multidisciplinary Digital Publishing Institute (MDPI), 1, 9.09%
Multidisciplinary Digital Publishing Institute (MDPI)
1 publication, 9.09%
American Institute of Physics (AIP)
American Institute of Physics (AIP), 1, 9.09%
American Institute of Physics (AIP)
1 publication, 9.09%
1
2
  • We do not take into account publications that without a DOI.
  • Statistics recalculated only for publications connected to researchers, organizations and labs registered on the platform.
  • Statistics recalculated weekly.
Metrics
Share
Cite this
GOST |
Cite this
GOST Copy
Yudin D., Slavioglo D. Usage of fully convolutional network with clustering for traffic light detection // 2018 7th Mediterranean Conference on Embedded Computing (MECO). 2018.
GOST all authors (up to 50) Copy
Yudin D., Slavioglo D. Usage of fully convolutional network with clustering for traffic light detection // 2018 7th Mediterranean Conference on Embedded Computing (MECO). 2018.
RIS |
Cite this
RIS Copy
TY - CPAPER
DO - 10.1109/MECO.2018.8406049
UR - https://doi.org/10.1109%2FMECO.2018.8406049
TI - Usage of fully convolutional network with clustering for traffic light detection
T2 - 2018 7th Mediterranean Conference on Embedded Computing (MECO)
AU - Yudin, D.
AU - Slavioglo, Dmitry
PY - 2018
DA - 2018/06/01 00:00:00
PB - IEEE
ER -
BibTex
Cite this
BibTex Copy
@inproceedings{2018_Yudin,
author = {D. Yudin and Dmitry Slavioglo},
title = {Usage of fully convolutional network with clustering for traffic light detection},
year = {2018},
month = {jun},
publisher = {IEEE}
}
Found error?