Automatic recognition of Colombian car license plates using convolutional neural networks and Chars74k database
A methodology for the automatic recognition of Colombian car license plates using convolutional neural networks is proposed. One of the biggest challenges when using onvolutional neural network is the demand for large amounts of samples for training. In this work, we show that if we do not have enou...
- Autores:
-
Arroyo-Pérez, D E
Álvarez-Canchila, O I
Patiño-Saucedo, A
Rostro González, H
Patiño Vanegas, Alberto
- Tipo de recurso:
- Fecha de publicación:
- 2020
- Institución:
- Universidad Tecnológica de Bolívar
- Repositorio:
- Repositorio Institucional UTB
- Idioma:
- eng
- OAI Identifier:
- oai:repositorio.utb.edu.co:20.500.12585/9384
- Acceso en línea:
- https://hdl.handle.net/20.500.12585/9384
- Palabra clave:
- Character Recognition
Tesseract
Template Matching
Convolution
Database systems
Image processing
License plates (automobile)
Colombians
Large amounts
Vehicle license plates
Convolutional neural networks
- Rights
- openAccess
- License
- http://creativecommons.org/licenses/by-nc/4.0/
id |
UTB2_2b7f411f74744c7780e1a473ee4d11f4 |
---|---|
oai_identifier_str |
oai:repositorio.utb.edu.co:20.500.12585/9384 |
network_acronym_str |
UTB2 |
network_name_str |
Repositorio Institucional UTB |
repository_id_str |
|
dc.title.spa.fl_str_mv |
Automatic recognition of Colombian car license plates using convolutional neural networks and Chars74k database |
title |
Automatic recognition of Colombian car license plates using convolutional neural networks and Chars74k database |
spellingShingle |
Automatic recognition of Colombian car license plates using convolutional neural networks and Chars74k database Character Recognition Tesseract Template Matching Convolution Database systems Image processing License plates (automobile) Colombians Large amounts Vehicle license plates Convolutional neural networks |
title_short |
Automatic recognition of Colombian car license plates using convolutional neural networks and Chars74k database |
title_full |
Automatic recognition of Colombian car license plates using convolutional neural networks and Chars74k database |
title_fullStr |
Automatic recognition of Colombian car license plates using convolutional neural networks and Chars74k database |
title_full_unstemmed |
Automatic recognition of Colombian car license plates using convolutional neural networks and Chars74k database |
title_sort |
Automatic recognition of Colombian car license plates using convolutional neural networks and Chars74k database |
dc.creator.fl_str_mv |
Arroyo-Pérez, D E Álvarez-Canchila, O I Patiño-Saucedo, A Rostro González, H Patiño Vanegas, Alberto |
dc.contributor.author.none.fl_str_mv |
Arroyo-Pérez, D E Álvarez-Canchila, O I Patiño-Saucedo, A Rostro González, H Patiño Vanegas, Alberto |
dc.subject.keywords.spa.fl_str_mv |
Character Recognition Tesseract Template Matching Convolution Database systems Image processing License plates (automobile) Colombians Large amounts Vehicle license plates Convolutional neural networks |
topic |
Character Recognition Tesseract Template Matching Convolution Database systems Image processing License plates (automobile) Colombians Large amounts Vehicle license plates Convolutional neural networks |
description |
A methodology for the automatic recognition of Colombian car license plates using convolutional neural networks is proposed. One of the biggest challenges when using onvolutional neural network is the demand for large amounts of samples for training. In this work, we show that if we do not have enough images of vehicle license plates to carry out the training, we can complement it with databases of letters and numbers that are not extracted from cars. The network was trained with the Chars74k database and images of characters extracted from plates of Colombian automobiles. The Chars74k contains approximately 74000 images of all the letters of the Spanish alphabet and all digits from 0 to 9. From chars74k database we have chosen 33849, because the Colombian plates have only uppercase letters and digits. Only 3549 (about 10% of the total) images of characters extracted manually from plates of Colombian automobiles were added. At the input of the convolutional neural network, 70% of the images were used for training, 20% for validation and 10% for testing and the resulting validation accuracy was above 99%. By making a preliminary test on Colombian plates never before used in training, a percentage of correctly recognized plates above 98% was achieved. |
publishDate |
2020 |
dc.date.accessioned.none.fl_str_mv |
2020-09-10T21:24:09Z |
dc.date.available.none.fl_str_mv |
2020-09-10T21:24:09Z |
dc.date.issued.none.fl_str_mv |
2020 |
dc.date.submitted.none.fl_str_mv |
2020-09-09 |
dc.type.coarversion.fl_str_mv |
http://purl.org/coar/version/c_970fb48d4fbd8a85 |
dc.type.coar.fl_str_mv |
http://purl.org/coar/resource_type/c_8544 |
dc.type.driver.spa.fl_str_mv |
info:eu-repo/semantics/lecture |
dc.type.hasversion.spa.fl_str_mv |
info:eu-repo/semantics/publishedVersion |
dc.type.spa.spa.fl_str_mv |
Otro |
status_str |
publishedVersion |
dc.identifier.citation.spa.fl_str_mv |
Arroyo-Pérez, D. E., Alvarez-Canchila, O. I., Patĩo-Saucedo, A., Rostro González, H., & Patĩo-Vanegas, A. (2020). Automatic recognition of colombian car license plates using convolutional neural networks and Chars74k database. Paper presented at the Journal of Physics: Conference Series, , 1547(1) doi:10.1088/1742-6596/1547/1/012024 |
dc.identifier.uri.none.fl_str_mv |
https://hdl.handle.net/20.500.12585/9384 |
dc.identifier.doi.none.fl_str_mv |
10.1088/1742-6596/1547/1/012024 |
dc.identifier.instname.spa.fl_str_mv |
Universidad Tecnológica de Bolívar |
dc.identifier.reponame.spa.fl_str_mv |
Repositorio Universidad Tecnológica de Bolívar |
identifier_str_mv |
Arroyo-Pérez, D. E., Alvarez-Canchila, O. I., Patĩo-Saucedo, A., Rostro González, H., & Patĩo-Vanegas, A. (2020). Automatic recognition of colombian car license plates using convolutional neural networks and Chars74k database. Paper presented at the Journal of Physics: Conference Series, , 1547(1) doi:10.1088/1742-6596/1547/1/012024 10.1088/1742-6596/1547/1/012024 Universidad Tecnológica de Bolívar Repositorio Universidad Tecnológica de Bolívar |
url |
https://hdl.handle.net/20.500.12585/9384 |
dc.language.iso.spa.fl_str_mv |
eng |
language |
eng |
dc.rights.coar.fl_str_mv |
http://purl.org/coar/access_right/c_abf2 |
dc.rights.uri.*.fl_str_mv |
http://creativecommons.org/licenses/by-nc/4.0/ |
dc.rights.accessrights.spa.fl_str_mv |
info:eu-repo/semantics/openAccess |
dc.rights.cc.*.fl_str_mv |
Atribución-NoComercial 4.0 Internacional |
rights_invalid_str_mv |
http://creativecommons.org/licenses/by-nc/4.0/ Atribución-NoComercial 4.0 Internacional http://purl.org/coar/access_right/c_abf2 |
eu_rights_str_mv |
openAccess |
dc.format.extent.none.fl_str_mv |
8 páginas |
dc.format.mimetype.spa.fl_str_mv |
application/pdf |
dc.coverage.spatial.none.fl_str_mv |
Colombia |
dc.publisher.place.spa.fl_str_mv |
Cartagena de Indias |
dc.source.spa.fl_str_mv |
Journal of Physics: Conference Series 1547 (2020) 012024 |
institution |
Universidad Tecnológica de Bolívar |
bitstream.url.fl_str_mv |
https://repositorio.utb.edu.co/bitstream/20.500.12585/9384/1/33.pdf https://repositorio.utb.edu.co/bitstream/20.500.12585/9384/2/license_rdf https://repositorio.utb.edu.co/bitstream/20.500.12585/9384/3/license.txt https://repositorio.utb.edu.co/bitstream/20.500.12585/9384/6/33.pdf.txt https://repositorio.utb.edu.co/bitstream/20.500.12585/9384/7/33.pdf.jpg |
bitstream.checksum.fl_str_mv |
e350002b0b36376d27b0e2e15d1bbe19 24013099e9e6abb1575dc6ce0855efd5 e20ad307a1c5f3f25af9304a7a7c86b6 4522ad2fe888f378769cc0b1228280ff 7b149af636e3d769f627db1209c17b6a |
bitstream.checksumAlgorithm.fl_str_mv |
MD5 MD5 MD5 MD5 MD5 |
repository.name.fl_str_mv |
Repositorio Institucional UTB |
repository.mail.fl_str_mv |
repositorioutb@utb.edu.co |
_version_ |
1814021688979881984 |
spelling |
Arroyo-Pérez, D E5af0863a-702e-4e04-aa7e-38bd47293639Álvarez-Canchila, O I4aa2cb07-d3ce-498e-bca2-6501469b0abaPatiño-Saucedo, Ac3a33eee-85f4-42c1-b646-2daa28d3a1cbRostro González, He0f28ea6-3fc7-4ad2-ade9-243a3fb477c4Patiño Vanegas, Alberto1466c16d-4b5e-49f0-8892-894e2ae6e66eColombia2020-09-10T21:24:09Z2020-09-10T21:24:09Z20202020-09-09Arroyo-Pérez, D. E., Alvarez-Canchila, O. I., Patĩo-Saucedo, A., Rostro González, H., & Patĩo-Vanegas, A. (2020). Automatic recognition of colombian car license plates using convolutional neural networks and Chars74k database. Paper presented at the Journal of Physics: Conference Series, , 1547(1) doi:10.1088/1742-6596/1547/1/012024https://hdl.handle.net/20.500.12585/938410.1088/1742-6596/1547/1/012024Universidad Tecnológica de BolívarRepositorio Universidad Tecnológica de BolívarA methodology for the automatic recognition of Colombian car license plates using convolutional neural networks is proposed. One of the biggest challenges when using onvolutional neural network is the demand for large amounts of samples for training. In this work, we show that if we do not have enough images of vehicle license plates to carry out the training, we can complement it with databases of letters and numbers that are not extracted from cars. The network was trained with the Chars74k database and images of characters extracted from plates of Colombian automobiles. The Chars74k contains approximately 74000 images of all the letters of the Spanish alphabet and all digits from 0 to 9. From chars74k database we have chosen 33849, because the Colombian plates have only uppercase letters and digits. Only 3549 (about 10% of the total) images of characters extracted manually from plates of Colombian automobiles were added. At the input of the convolutional neural network, 70% of the images were used for training, 20% for validation and 10% for testing and the resulting validation accuracy was above 99%. By making a preliminary test on Colombian plates never before used in training, a percentage of correctly recognized plates above 98% was achieved.8 páginasapplication/pdfenghttp://creativecommons.org/licenses/by-nc/4.0/info:eu-repo/semantics/openAccessAtribución-NoComercial 4.0 Internacionalhttp://purl.org/coar/access_right/c_abf2Journal of Physics: Conference Series 1547 (2020) 012024Automatic recognition of Colombian car license plates using convolutional neural networks and Chars74k databaseinfo:eu-repo/semantics/lectureinfo:eu-repo/semantics/publishedVersionOtrohttp://purl.org/coar/version/c_970fb48d4fbd8a85http://purl.org/coar/resource_type/c_8544Character RecognitionTesseractTemplate MatchingConvolutionDatabase systemsImage processingLicense plates (automobile)ColombiansLarge amountsVehicle license platesConvolutional neural networksCartagena de IndiasPúblico generalShivakumara P, Tang D, Asadzadehkaljahi M, Lu T, Pal U and Hossein Anisi M 2018 CAAI Transactions on Intelligence Technology 3(3) 169Halim S, Zulkifli M and Zulkipli M 2019 Journal of Physics: Conference Series 1358(012084) 1Astawa I, Bawa I, Asri S and Kariati N 2019 International Journal of Scientific and Technology Research 8(11) 481Du S, Ibrahim M, Shehata M and Badawy W 2013 IEEE Transactions on Circuits and Systems for Video Technology 23(2) 311Lecun Y, Bengio Y and Hinton G 2015 Nature 521(7553) 436Calderon J, Vargas J and Perez-Ruiz A 2017 License plate recognition for colombian private vehicles based on an embedded system using the zedboard IEEE Colombian Conference on Robotics and Automation (CCRA) (Bogotá: IEEE)De Campos T E, Babu B R, Varma M et al. 2009 Character recognition in natural images Proceedings of the International Conference on Computer Vision Theory and Applications (VISAPP) vol 2 (Lisbon: INSTICC Press)Mallat S 2016 Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences 374(2065) 1Schmidhuber J 2015 Neural Networks 61 85Rumelhart D E, Durbin R, Golden R and Chauvin Y 1995 Backpropagation: The basic theory Developments in Connectionist Theory. Backpropagation: Theory, Architectures, and Applications (New Jersey: Lawrence Erlbaum Associates Inc.) pp 1–34Gonzalez R C and Woods R E 2017 Digital Image Processing (New York: Pearson)Patino-Saucedo A, Rostro-Gonzalez H and Conradt J 2018 Tropical fruits classification using an alexnet-type convolutional neural network and image augmentation International Conference on Neural Information Processing (Cambodia: Springer) pp 371–379Bisong E 2019 Google colaboratory Building Machine Learning and Deep Learning Models on Google Cloud Platform: A Comprehensive Guide for Beginners (Berkeley: Apress) pp 59–64Wu S, Zhai W and Cao Y 2019 IET Image Processing 13(14) 2744Kessentini Y, Besbes M, Ammar S and Chabbouh A 2019 Expert Systems with Applications 136 159http://purl.org/coar/resource_type/c_c94fORIGINAL33.pdf33.pdfPonenciaapplication/pdf5660280https://repositorio.utb.edu.co/bitstream/20.500.12585/9384/1/33.pdfe350002b0b36376d27b0e2e15d1bbe19MD51CC-LICENSElicense_rdflicense_rdfapplication/rdf+xml; charset=utf-8914https://repositorio.utb.edu.co/bitstream/20.500.12585/9384/2/license_rdf24013099e9e6abb1575dc6ce0855efd5MD52LICENSElicense.txtlicense.txttext/plain; charset=utf-83182https://repositorio.utb.edu.co/bitstream/20.500.12585/9384/3/license.txte20ad307a1c5f3f25af9304a7a7c86b6MD53TEXT33.pdf.txt33.pdf.txtExtracted texttext/plain20205https://repositorio.utb.edu.co/bitstream/20.500.12585/9384/6/33.pdf.txt4522ad2fe888f378769cc0b1228280ffMD56THUMBNAIL33.pdf.jpg33.pdf.jpgGenerated Thumbnailimage/jpeg34563https://repositorio.utb.edu.co/bitstream/20.500.12585/9384/7/33.pdf.jpg7b149af636e3d769f627db1209c17b6aMD5720.500.12585/9384oai:repositorio.utb.edu.co:20.500.12585/93842023-05-25 10:02:21.499Repositorio Institucional UTBrepositorioutb@utb.edu.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 |