A modular neural network classifier for the recognition of occluded characters in automatic license plate reading
(2002) 5th International Conference on Computational Intelligent Systems for Applied Research (FLINS) p.363-372- 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.
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/603909
- author
- Nijhuis, J A G ; Broersma, A and Spaanenburg, Lambert LU
- publishing date
- 2002
- type
- Chapter in Book/Report/Conference proceeding
- publication status
- published
- subject
- host publication
- COMPUTATIONAL INTELLIGENT SYSTEMS FOR APPLIED RESEARCH : Proceedings of the 5th International FLINS Conference
- editor
- Da Ruan, Pierre D'hondt and Kerre, Etienne E
- pages
- 363 - 372
- publisher
- World Scientific Publishing
- conference name
- 5th International Conference on Computational Intelligent Systems for Applied Research (FLINS)
- conference location
- Gent, Belgium
- conference dates
- 2002-09-16 - 2002-09-18
- ISBN
- 978-981-238-066-1
- language
- English
- LU publication?
- no
- id
- d8939333-df10-45d5-ac94-5e84895e29ae (old id 603909)
- date added to LUP
- 2016-04-04 10:31:04
- date last changed
- 2021-01-25 10:54:20
@inproceedings{d8939333-df10-45d5-ac94-5e84895e29ae, 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.}}, author = {{Nijhuis, J A G and Broersma, A and Spaanenburg, Lambert}}, booktitle = {{COMPUTATIONAL INTELLIGENT SYSTEMS FOR APPLIED RESEARCH : Proceedings of the 5th International FLINS Conference}}, editor = {{Da Ruan, Pierre D'hondt and Kerre, Etienne E}}, isbn = {{978-981-238-066-1}}, language = {{eng}}, pages = {{363--372}}, publisher = {{World Scientific Publishing}}, title = {{A modular neural network classifier for the recognition of occluded characters in automatic license plate reading}}, year = {{2002}}, }