Polish Road Signs Detection and Classification System Based on Sign Sketches and ConvNet

Łukasz Chechliński , Bartłomiej Chechliński


In this paper, we present a novel approach to detection and classification of road traffic signs. Detection and classification is performed simultaneously by the Deep Convolutional Neural Network, based on the architecture of VGG Net. Classifier is trained with the usage of sign sketches, obtained directly from the Polish Highway Code. All 169 simple signs are used. The system was tested on 100 images obtained from Google Street View. The re-view of related work shows that our system does not reach the state-of-the-art results yet, but it is much easily scalable and adaptable to the new high-way codes.
Author Łukasz Chechliński (FM / IACR)
Łukasz Chechliński,,
- The Institute of Automatic Control and Robotics
, Bartłomiej Chechliński - [Faculty of Mathematics and Information Science WUT Warsaw Poland]
Bartłomiej Chechliński,,
- Faculty of Mathematics and Information Science WUT Warsaw Poland
Publication size in sheets0.5
Book Březina Tomáš, Jabłoński Ryszard (eds.): Mechatronics 2017: Recent Technological and Scientific Advances, Advances in Intelligent Systems and Computing, vol. 644, 2018, Springer, ISBN 978-3-319-65959-6, 753 p., DOI:10.1007/978-3-319-65960-2
Keywords in Englishtraffic signs, deep learning, ConvNet, synthetic training base, sign detection
URL https://link.springer.com/chapter/10.1007/978-3-319-65960-2_67
Languageen angielski
Score (nominal)20
Score sourcepublisherList
ScoreMinisterial score = 20.0, 10-01-2020, ChapterFromConference
Citation count*
Share Share

Get link to the record

* presented citation count is obtained through Internet information analysis and it is close to the number calculated by the Publish or Perish system.
Are you sure?