A bio-inspired quaternion local phase CNN layer with contrast invariance and linear sensitivity to rotation angles

Deep learning models have been particularly successful with image recognition using Convolutional Neural Networks (CNN). However, the learning of a contrast invariance and rotation equivariance response may fail even with very deep CNNs or by large data augmentations in training. We were inspired by...

Descripción completa

Detalles Bibliográficos
Autores: Moya Sánchez, Eduardo Ulises, Xambó Descamps, Sebastián|||0000-0001-5056-9818, Sánchez Pérez, Abraham, Salazar Colores, Sebastián, Martínez Ortega, Jorge, Cortés García, Claudio Ulises|||0000-0003-0192-3096
Tipo de recurso: artículo
Fecha de publicación:2020
País:España
Institución:Universitat Politècnica de Catalunya (UPC)
Repositorio:UPCommons. Portal del coneixement obert de la UPC
Idioma:inglés
OAI Identifier:oai:upcommons.upc.edu:2117/422915
Acceso en línea:https://hdl.handle.net/2117/422915
https://dx.doi.org/10.1016/j.patrec.2019.12.001
Access Level:acceso abierto
Palabra clave:Convolution
Deep learning
Image recognition
Rotation
Multilayer neural networks
Mammals
Àrees temàtiques de la UPC::Informàtica::Intel·ligència artificial::Aprenentatge automàtic
Àrees temàtiques de la UPC::Enginyeria de la telecomunicació::Processament del senyal::Processament de la imatge i del senyal vídeo
id ES_bb3a16faebc8a286b2201a8984edb14f
oai_identifier_str oai:upcommons.upc.edu:2117/422915
network_acronym_str ES
network_name_str España
repository_id_str
spelling A bio-inspired quaternion local phase CNN layer with contrast invariance and linear sensitivity to rotation anglesMoya Sánchez, Eduardo UlisesXambó Descamps, Sebastián|||0000-0001-5056-9818Sánchez Pérez, AbrahamSalazar Colores, SebastiánMartínez Ortega, JorgeCortés García, Claudio Ulises|||0000-0003-0192-3096ConvolutionDeep learningImage recognitionRotationMultilayer neural networksMammalsÀrees temàtiques de la UPC::Informàtica::Intel·ligència artificial::Aprenentatge automàticÀrees temàtiques de la UPC::Enginyeria de la telecomunicació::Processament del senyal::Processament de la imatge i del senyal vídeoDeep learning models have been particularly successful with image recognition using Convolutional Neural Networks (CNN). However, the learning of a contrast invariance and rotation equivariance response may fail even with very deep CNNs or by large data augmentations in training. We were inspired by the V1 visual features of the mammalian visual system to emulate as much as possible the early visual system and add more invariant capacities to the CNN. We present a new quaternion local phase convolutional neural network layer encoding three local phases. We present two experimental setups: An image classification task with four contrast levels, and a linear regression task that predicts the rotation angle of an image. In sum, we obtain new patterns and feature representations for deep learning, which capture illumination invariance and a linear response to rotation angles.The authors would like to thank to CONACYT and the Barcelona supercomputing Center. Sebastián Salazar-Colores (CVU 477758) would like to thank CONACYT (Consejo Nacional de Ciencia y Tecnología) for the financial support of his PhD studies under Scholarship 285651. Eduardo Ulises Moya-Sánchez, Jorge Martínez-Ortega and Ulises Cortés are members of the Sistema Nacional de Investigadores (SNI) de México.Peer ReviewedElsevier20202020-03-0120252025-01-28journal articlehttp://purl.org/coar/resource_type/c_6501AMhttp://purl.org/coar/version/c_ab4af688f83e57aainfo:eu-repo/semantics/articleapplication/pdfhttps://hdl.handle.net/2117/422915https://dx.doi.org/10.1016/j.patrec.2019.12.001reponame:UPCommons. Portal del coneixement obert de la UPCinstname:Universitat Politècnica de Catalunya (UPC)Inglésengopen accesshttp://purl.org/coar/access_right/c_abf2Attribution-NonCommercial-NoDerivatives 4.0 Internationalhttp://creativecommons.org/licenses/by-nc-nd/4.0/info:eu-repo/semantics/openAccessoai:upcommons.upc.edu:2117/4229152026-05-27T15:37:01Z
dc.title.none.fl_str_mv A bio-inspired quaternion local phase CNN layer with contrast invariance and linear sensitivity to rotation angles
title A bio-inspired quaternion local phase CNN layer with contrast invariance and linear sensitivity to rotation angles
spellingShingle A bio-inspired quaternion local phase CNN layer with contrast invariance and linear sensitivity to rotation angles
Moya Sánchez, Eduardo Ulises
Convolution
Deep learning
Image recognition
Rotation
Multilayer neural networks
Mammals
Àrees temàtiques de la UPC::Informàtica::Intel·ligència artificial::Aprenentatge automàtic
Àrees temàtiques de la UPC::Enginyeria de la telecomunicació::Processament del senyal::Processament de la imatge i del senyal vídeo
title_short A bio-inspired quaternion local phase CNN layer with contrast invariance and linear sensitivity to rotation angles
title_full A bio-inspired quaternion local phase CNN layer with contrast invariance and linear sensitivity to rotation angles
title_fullStr A bio-inspired quaternion local phase CNN layer with contrast invariance and linear sensitivity to rotation angles
title_full_unstemmed A bio-inspired quaternion local phase CNN layer with contrast invariance and linear sensitivity to rotation angles
title_sort A bio-inspired quaternion local phase CNN layer with contrast invariance and linear sensitivity to rotation angles
dc.creator.none.fl_str_mv Moya Sánchez, Eduardo Ulises
Xambó Descamps, Sebastián|||0000-0001-5056-9818
Sánchez Pérez, Abraham
Salazar Colores, Sebastián
Martínez Ortega, Jorge
Cortés García, Claudio Ulises|||0000-0003-0192-3096
author Moya Sánchez, Eduardo Ulises
author_facet Moya Sánchez, Eduardo Ulises
Xambó Descamps, Sebastián|||0000-0001-5056-9818
Sánchez Pérez, Abraham
Salazar Colores, Sebastián
Martínez Ortega, Jorge
Cortés García, Claudio Ulises|||0000-0003-0192-3096
author_role author
author2 Xambó Descamps, Sebastián|||0000-0001-5056-9818
Sánchez Pérez, Abraham
Salazar Colores, Sebastián
Martínez Ortega, Jorge
Cortés García, Claudio Ulises|||0000-0003-0192-3096
author2_role author
author
author
author
author
dc.subject.none.fl_str_mv Convolution
Deep learning
Image recognition
Rotation
Multilayer neural networks
Mammals
Àrees temàtiques de la UPC::Informàtica::Intel·ligència artificial::Aprenentatge automàtic
Àrees temàtiques de la UPC::Enginyeria de la telecomunicació::Processament del senyal::Processament de la imatge i del senyal vídeo
topic Convolution
Deep learning
Image recognition
Rotation
Multilayer neural networks
Mammals
Àrees temàtiques de la UPC::Informàtica::Intel·ligència artificial::Aprenentatge automàtic
Àrees temàtiques de la UPC::Enginyeria de la telecomunicació::Processament del senyal::Processament de la imatge i del senyal vídeo
description Deep learning models have been particularly successful with image recognition using Convolutional Neural Networks (CNN). However, the learning of a contrast invariance and rotation equivariance response may fail even with very deep CNNs or by large data augmentations in training. We were inspired by the V1 visual features of the mammalian visual system to emulate as much as possible the early visual system and add more invariant capacities to the CNN. We present a new quaternion local phase convolutional neural network layer encoding three local phases. We present two experimental setups: An image classification task with four contrast levels, and a linear regression task that predicts the rotation angle of an image. In sum, we obtain new patterns and feature representations for deep learning, which capture illumination invariance and a linear response to rotation angles.
publishDate 2020
dc.date.none.fl_str_mv 2020
2020-03-01
2025
2025-01-28
dc.type.none.fl_str_mv journal article
http://purl.org/coar/resource_type/c_6501
AM
http://purl.org/coar/version/c_ab4af688f83e57aa
dc.type.openaire.fl_str_mv info:eu-repo/semantics/article
format article
dc.identifier.none.fl_str_mv https://hdl.handle.net/2117/422915
https://dx.doi.org/10.1016/j.patrec.2019.12.001
url https://hdl.handle.net/2117/422915
https://dx.doi.org/10.1016/j.patrec.2019.12.001
dc.language.none.fl_str_mv Inglés
eng
language_invalid_str_mv Inglés
language eng
dc.rights.none.fl_str_mv open access
http://purl.org/coar/access_right/c_abf2
Attribution-NonCommercial-NoDerivatives 4.0 International
http://creativecommons.org/licenses/by-nc-nd/4.0/
dc.rights.openaire.fl_str_mv info:eu-repo/semantics/openAccess
rights_invalid_str_mv open access
http://purl.org/coar/access_right/c_abf2
Attribution-NonCommercial-NoDerivatives 4.0 International
http://creativecommons.org/licenses/by-nc-nd/4.0/
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
dc.publisher.none.fl_str_mv Elsevier
publisher.none.fl_str_mv Elsevier
dc.source.none.fl_str_mv reponame:UPCommons. Portal del coneixement obert de la UPC
instname:Universitat Politècnica de Catalunya (UPC)
instname_str Universitat Politècnica de Catalunya (UPC)
reponame_str UPCommons. Portal del coneixement obert de la UPC
collection UPCommons. Portal del coneixement obert de la UPC
repository.name.fl_str_mv
repository.mail.fl_str_mv
_version_ 1869418007128702976
score 15,81155

Similares en LA Referencia