A modular neural network classifier for the recognition of occluded characters in automatic license plate reading

Forskningsoutput: Kapitel i bok/rapport/Conference proceedingKonferenspaper i proceeding

Abstract

Occlusion is the most common reason for lowered recognition yield in free-flow license-plate reading systems. (Non-)occluded characters can readily be learned in separate neural networks but not together. Even a small proportion of occluded characters in the training set will already significantly reduce the overall recognition yield. This paper shows that a modular network can handle a realistic mixture of (non-) occluded characters with a 99.8% recognition yield per character.

Detaljer

Författare
  • J A G Nijhuis
  • A Broersma
  • Lambert Spaanenburg
Externa organisationer
  • External Organization - Unknown
Forskningsområden

Ämnesklassifikation (UKÄ) – OBLIGATORISK

  • Elektroteknik och elektronik
Originalspråkengelska
Titel på värdpublikationCOMPUTATIONAL INTELLIGENT SYSTEMS FOR APPLIED RESEARCH : Proceedings of the 5th International FLINS Conference
RedaktörerPierre D'hondt Da Ruan, Etienne E Kerre
FörlagWorld Scientific
Sidor363-372
ISBN (tryckt)978-981-238-066-1
StatusPublished - 2002
PublikationskategoriForskning
Peer review utfördJa
Externt publiceradJa
Evenemang5th International Conference on Computational Intelligent Systems for Applied Research (FLINS) - Gent, Belgien
Varaktighet: 2002 sep 162002 sep 18

Konferens

Konferens5th International Conference on Computational Intelligent Systems for Applied Research (FLINS)
LandBelgien
OrtGent
Period2002/09/162002/09/18