Traffic sign detection and recognition system

Sergio Escalera*, Xavier Baró, Oriol Pujol, Jordi Vitrià, Petia Radeva

*Corresponding author for this work

Research output: Contribution to book/anthology/report/conference proceedingBook chapterResearchpeer-review

1 Citation (Scopus)

Abstract

Detection and recognition steps are the key elements of a full traffic sign system. In this chapter, a complete description of a complete traffic sign system and how the component pieces are glued together are provided. Starting with the stereo acquisition module, data are fed into an attentional cascade detection algorithm. This step is performed separately for each video source and the coherence of the detections for the stereo pair is checked. As a result the location of the traffic sign and its size if obtained. The detected signs are normalized taking into account their shape. Then, they are given to the Forest-ECOC classification stage. All stages are validated individually and as a whole providing competitive results with respect to the state-of-the-art techniques.

Original languageEnglish
Title of host publicationSpringerBriefs in Computer Science
Number of pages14
PublisherSpringer VS
Publication date2011
Edition9781447122449
Pages81-94
DOIs
Publication statusPublished - 2011
Externally publishedYes
SeriesSpringerBriefs in Computer Science
Number9781447122449
Volume0
ISSN2191-5768

Bibliographical note

Publisher Copyright:
© Sergio Escalera 2011.

Keywords

  • Geovan
  • Mobile mapping systems
  • Model fitting
  • Stereo association
  • Traffic sign data set
  • Traffic sign recognition

Fingerprint

Dive into the research topics of 'Traffic sign detection and recognition system'. Together they form a unique fingerprint.

Cite this