Method for classifying images in databases through deep convolutional networks
Since 2006, deep structured learning, or more commonly called deep learning or hierarchical learning, has become a new area of research in machine learning. In recent years, techniques developed from deep learning research have impacted on a wide range of information and particularly image processin...
- Autores:
-
Varela, Noel
Comas-González, Zoe
Ternera-Muñoz, Yesith R
Esmeral-Romero, Ernesto F
Lizardo Zelaya, Nelson Alberto
- Tipo de recurso:
- Article of journal
- Fecha de publicación:
- 2020
- Institución:
- Corporación Universidad de la Costa
- Repositorio:
- REDICUC - Repositorio CUC
- Idioma:
- eng
- OAI Identifier:
- oai:repositorio.cuc.edu.co:11323/7660
- Acceso en línea:
- https://hdl.handle.net/11323/7660
https://doi.org/10.1016/j.procs.2020.07.022
https://repositorio.cuc.edu.co/
- Palabra clave:
- Dynamic training
Deep convolutional networks
Image classification
- Rights
- openAccess
- License
- CC0 1.0 Universal
| id |
RCUC2_eb49d42f0b4922c469462accf1de247b |
|---|---|
| oai_identifier_str |
oai:repositorio.cuc.edu.co:11323/7660 |
| network_acronym_str |
RCUC2 |
| network_name_str |
REDICUC - Repositorio CUC |
| repository_id_str |
|
| dc.title.spa.fl_str_mv |
Method for classifying images in databases through deep convolutional networks |
| title |
Method for classifying images in databases through deep convolutional networks |
| spellingShingle |
Method for classifying images in databases through deep convolutional networks Dynamic training Deep convolutional networks Image classification |
| title_short |
Method for classifying images in databases through deep convolutional networks |
| title_full |
Method for classifying images in databases through deep convolutional networks |
| title_fullStr |
Method for classifying images in databases through deep convolutional networks |
| title_full_unstemmed |
Method for classifying images in databases through deep convolutional networks |
| title_sort |
Method for classifying images in databases through deep convolutional networks |
| dc.creator.fl_str_mv |
Varela, Noel Comas-González, Zoe Ternera-Muñoz, Yesith R Esmeral-Romero, Ernesto F Lizardo Zelaya, Nelson Alberto |
| dc.contributor.author.spa.fl_str_mv |
Varela, Noel Comas-González, Zoe Ternera-Muñoz, Yesith R Esmeral-Romero, Ernesto F Lizardo Zelaya, Nelson Alberto |
| dc.subject.spa.fl_str_mv |
Dynamic training Deep convolutional networks Image classification |
| topic |
Dynamic training Deep convolutional networks Image classification |
| description |
Since 2006, deep structured learning, or more commonly called deep learning or hierarchical learning, has become a new area of research in machine learning. In recent years, techniques developed from deep learning research have impacted on a wide range of information and particularly image processing studies, within traditional and new fields, including key aspects of machine learning and artificial intelligence. This paper proposes an alternative scheme for training data management in CNNs, consisting of selective-adaptive data sampling. By means of experiments with the CIFAR10 database for image classification. |
| publishDate |
2020 |
| dc.date.issued.none.fl_str_mv |
2020 |
| dc.date.accessioned.none.fl_str_mv |
2021-01-05T21:45:46Z |
| dc.date.available.none.fl_str_mv |
2021-01-05T21:45:46Z |
| dc.type.spa.fl_str_mv |
Artículo de revista |
| dc.type.coar.fl_str_mv |
http://purl.org/coar/resource_type/c_2df8fbb1 |
| dc.type.coar.spa.fl_str_mv |
http://purl.org/coar/resource_type/c_6501 |
| dc.type.content.spa.fl_str_mv |
Text |
| dc.type.driver.spa.fl_str_mv |
info:eu-repo/semantics/article |
| dc.type.redcol.spa.fl_str_mv |
http://purl.org/redcol/resource_type/ART |
| dc.type.version.spa.fl_str_mv |
info:eu-repo/semantics/acceptedVersion |
| format |
http://purl.org/coar/resource_type/c_6501 |
| status_str |
acceptedVersion |
| dc.identifier.issn.spa.fl_str_mv |
1877-0509 |
| dc.identifier.uri.spa.fl_str_mv |
https://hdl.handle.net/11323/7660 |
| dc.identifier.doi.spa.fl_str_mv |
https://doi.org/10.1016/j.procs.2020.07.022 |
| dc.identifier.instname.spa.fl_str_mv |
Corporación Universidad de la Costa |
| dc.identifier.reponame.spa.fl_str_mv |
REDICUC - Repositorio CUC |
| dc.identifier.repourl.spa.fl_str_mv |
https://repositorio.cuc.edu.co/ |
| identifier_str_mv |
1877-0509 Corporación Universidad de la Costa REDICUC - Repositorio CUC |
| url |
https://hdl.handle.net/11323/7660 https://doi.org/10.1016/j.procs.2020.07.022 https://repositorio.cuc.edu.co/ |
| dc.language.iso.none.fl_str_mv |
eng |
| language |
eng |
| dc.relation.references.spa.fl_str_mv |
[1] Singh, R., Khurana, R., Kushwaha, A. K. S., & Srivastava, R. (2020). Combining CNN streams of dynamic image and depth data for action recognition. Multimedia Systems, 1-10. [2] Mostafa, H., & Wang, X. (2019). Parameter efficient training of deep convolutional neural networks by dynamic sparse reparameterization. arXiv preprint arXiv:1902.05967. [3] Viloria, A., Angulo, M. G., Kamatkar, S. J., de la Hoz – Hernandez, J., Guiliany, J. G., Bilbao, O. R., & Hernandez-P, H. (2020). Prediction Rules in E-Learning Systems Using Genetic Programming. In Smart Innovation, Systems and Technologies (Vol. 164, pp. 55–63). Springer. https://doi.org/10.1007/978-981-32-9889-7_5. [4] Zheng, Q., Yang, M., Tian, X., Jiang, N., & Wang, D. (2020). A Full Stage Data Augmentation Method in Deep Convolutional Neural Network for Natural Image Classification. Discrete Dynamics in Nature and Society, 2020. [5] Haque, N., Reddy, N. D., & Krishna, K. M. (2017). Joint semantic and motion segmentation for dynamic scenes using deep convolutional networks. arXiv preprint arXiv:1704.08331. [6] Wang, H. M. X. (2019). Parameter Efficient Training of Deep Convolutional Neural Networks by Dynamic Sparse Reparameterization. arXiv preprint arXiv:1902.05967. [7] Tang, M., Liu, Y., & Durlofsky, L. J. (2020). A deep-learning-based surrogate model for data assimilation in dynamic subsurface flow problems. Journal of Computational Physics, 109456. [8] Yang, J., Liang, J., Shen, H., Wang, K., Rosin, P. L., & Yang, M. H. (2018). Dynamic match kernel with deep convolutional features for image retrieval. IEEE Transactions on Image Processing, 27(11), 5288-5302. [9] Popov, V., Shakev, N., Ahmed, S., & Toplaov, A. (2018, September). Recognition of Dynamic Targets using a Deep Convolutional Neural Network. In ANNA'18; Advances in Neural Networks and Applications 2018 (pp. 1-6). VDE. [10] Aimone, J. B., & Severa, W. M. (2017). Context-modulation of hippocampal dynamics and deep convolutional networks. arXiv preprint arXiv:1711.09876. [11] Nah, S., Hyun Kim, T., & Mu Lee, K. (2017). Deep multi-scale convolutional neural network for dynamic scene deblurring. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (pp. 3883-3891). [12] Manessi, F., Rozza, A., & Manzo, M. (2020). Dynamic graph convolutional networks. Pattern Recognition, 97, 107000. [13] Mo, S., Zhu, Y., Zabaras, N., Shi, X., & Wu, J. (2019). Deep convolutional encoder‐decoder networks for uncertainty quantification of dynamic multiphase flow in heterogeneous media. Water Resources Research, 55(1), 703-728. [14] Viloria, A., Varela, N., Lezama, O. B. P., Llinás, N. O., Flores, Y., Palma, H. H., … Marín-González, F. (2020). Classification of Digitized Documents Applying Neural Networks. In Lecture Notes in Electrical Engineering (Vol. 637, pp. 213–220). Springer. https://doi.org/10.1007/978-981-15-2612-1_20. [15] Shao, R., Lan, X., & Yuen, P. C. (2017, October). Deep convolutional dynamic texture learning with adaptive channel-discriminability for 3D mask face anti-spoofing. In 2017 IEEE International Joint Conference on Biometrics (IJCB) (pp. 748-755). IEEE. [16] Bak, C., Erdem, A., & Erdem, E. (2016). Two-stream convolutional networks for dynamic saliency prediction. arXiv preprint arXiv:1607.04730, 2(3), 6. |
| dc.rights.spa.fl_str_mv |
CC0 1.0 Universal |
| dc.rights.uri.spa.fl_str_mv |
http://creativecommons.org/publicdomain/zero/1.0/ |
| dc.rights.accessrights.spa.fl_str_mv |
info:eu-repo/semantics/openAccess |
| dc.rights.coar.spa.fl_str_mv |
http://purl.org/coar/access_right/c_abf2 |
| rights_invalid_str_mv |
CC0 1.0 Universal http://creativecommons.org/publicdomain/zero/1.0/ http://purl.org/coar/access_right/c_abf2 |
| eu_rights_str_mv |
openAccess |
| dc.format.mimetype.spa.fl_str_mv |
application/pdf |
| dc.publisher.spa.fl_str_mv |
Corporación Universidad de la Costa |
| dc.source.spa.fl_str_mv |
Procedia Computer Science |
| institution |
Corporación Universidad de la Costa |
| dc.source.url.spa.fl_str_mv |
https://www.sciencedirect.com/science/article/pii/S1877050920317014 |
| bitstream.url.fl_str_mv |
https://repositorio.cuc.edu.co/bitstreams/807d7e40-9468-47a3-a560-0e04ecdc534d/download https://repositorio.cuc.edu.co/bitstreams/4754bdec-6918-4c74-8c3d-4ecb85c9eee7/download https://repositorio.cuc.edu.co/bitstreams/846debed-f6c2-46c3-8e90-22f2b86c5afd/download https://repositorio.cuc.edu.co/bitstreams/9cef13a0-6159-45a0-bf5b-056c25f2915f/download https://repositorio.cuc.edu.co/bitstreams/8f0da30c-d362-487c-bdee-a426422af3ad/download |
| bitstream.checksum.fl_str_mv |
952555b8fcb24fd3faac787b9ed11e3a 42fd4ad1e89814f5e4a476b409eb708c e30e9215131d99561d40d6b0abbe9bad b21703a4d16891344ec3fbb0751fd591 b29a11585eb13b8743c87c0719ab8f47 |
| bitstream.checksumAlgorithm.fl_str_mv |
MD5 MD5 MD5 MD5 MD5 |
| repository.name.fl_str_mv |
Repositorio de la Universidad de la Costa CUC |
| repository.mail.fl_str_mv |
repdigital@cuc.edu.co |
| _version_ |
1811760679477051392 |
| spelling |
Varela, NoelComas-González, ZoeTernera-Muñoz, Yesith REsmeral-Romero, Ernesto FLizardo Zelaya, Nelson Alberto2021-01-05T21:45:46Z2021-01-05T21:45:46Z20201877-0509https://hdl.handle.net/11323/7660https://doi.org/10.1016/j.procs.2020.07.022Corporación Universidad de la CostaREDICUC - Repositorio CUChttps://repositorio.cuc.edu.co/Since 2006, deep structured learning, or more commonly called deep learning or hierarchical learning, has become a new area of research in machine learning. In recent years, techniques developed from deep learning research have impacted on a wide range of information and particularly image processing studies, within traditional and new fields, including key aspects of machine learning and artificial intelligence. This paper proposes an alternative scheme for training data management in CNNs, consisting of selective-adaptive data sampling. By means of experiments with the CIFAR10 database for image classification.Varela, NoelComas-González, ZoeTernera-Muñoz, Yesith REsmeral-Romero, Ernesto FLizardo Zelaya, Nelson Albertoapplication/pdfengCorporación Universidad de la CostaCC0 1.0 Universalhttp://creativecommons.org/publicdomain/zero/1.0/info:eu-repo/semantics/openAccesshttp://purl.org/coar/access_right/c_abf2Procedia Computer Sciencehttps://www.sciencedirect.com/science/article/pii/S1877050920317014Dynamic trainingDeep convolutional networksImage classificationMethod for classifying images in databases through deep convolutional networksArtículo de revistahttp://purl.org/coar/resource_type/c_6501http://purl.org/coar/resource_type/c_2df8fbb1Textinfo:eu-repo/semantics/articlehttp://purl.org/redcol/resource_type/ARTinfo:eu-repo/semantics/acceptedVersion[1] Singh, R., Khurana, R., Kushwaha, A. K. S., & Srivastava, R. (2020). Combining CNN streams of dynamic image and depth data for action recognition. Multimedia Systems, 1-10.[2] Mostafa, H., & Wang, X. (2019). Parameter efficient training of deep convolutional neural networks by dynamic sparse reparameterization. arXiv preprint arXiv:1902.05967.[3] Viloria, A., Angulo, M. G., Kamatkar, S. J., de la Hoz – Hernandez, J., Guiliany, J. G., Bilbao, O. R., & Hernandez-P, H. (2020). Prediction Rules in E-Learning Systems Using Genetic Programming. In Smart Innovation, Systems and Technologies (Vol. 164, pp. 55–63). Springer. https://doi.org/10.1007/978-981-32-9889-7_5.[4] Zheng, Q., Yang, M., Tian, X., Jiang, N., & Wang, D. (2020). A Full Stage Data Augmentation Method in Deep Convolutional Neural Network for Natural Image Classification. Discrete Dynamics in Nature and Society, 2020.[5] Haque, N., Reddy, N. D., & Krishna, K. M. (2017). Joint semantic and motion segmentation for dynamic scenes using deep convolutional networks. arXiv preprint arXiv:1704.08331.[6] Wang, H. M. X. (2019). Parameter Efficient Training of Deep Convolutional Neural Networks by Dynamic Sparse Reparameterization. arXiv preprint arXiv:1902.05967.[7] Tang, M., Liu, Y., & Durlofsky, L. J. (2020). A deep-learning-based surrogate model for data assimilation in dynamic subsurface flow problems. Journal of Computational Physics, 109456.[8] Yang, J., Liang, J., Shen, H., Wang, K., Rosin, P. L., & Yang, M. H. (2018). Dynamic match kernel with deep convolutional features for image retrieval. IEEE Transactions on Image Processing, 27(11), 5288-5302.[9] Popov, V., Shakev, N., Ahmed, S., & Toplaov, A. (2018, September). Recognition of Dynamic Targets using a Deep Convolutional Neural Network. In ANNA'18; Advances in Neural Networks and Applications 2018 (pp. 1-6). VDE.[10] Aimone, J. B., & Severa, W. M. (2017). Context-modulation of hippocampal dynamics and deep convolutional networks. arXiv preprint arXiv:1711.09876.[11] Nah, S., Hyun Kim, T., & Mu Lee, K. (2017). Deep multi-scale convolutional neural network for dynamic scene deblurring. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (pp. 3883-3891).[12] Manessi, F., Rozza, A., & Manzo, M. (2020). Dynamic graph convolutional networks. Pattern Recognition, 97, 107000.[13] Mo, S., Zhu, Y., Zabaras, N., Shi, X., & Wu, J. (2019). Deep convolutional encoder‐decoder networks for uncertainty quantification of dynamic multiphase flow in heterogeneous media. Water Resources Research, 55(1), 703-728.[14] Viloria, A., Varela, N., Lezama, O. B. P., Llinás, N. O., Flores, Y., Palma, H. H., … Marín-González, F. (2020). Classification of Digitized Documents Applying Neural Networks. In Lecture Notes in Electrical Engineering (Vol. 637, pp. 213–220). Springer. https://doi.org/10.1007/978-981-15-2612-1_20.[15] Shao, R., Lan, X., & Yuen, P. C. (2017, October). Deep convolutional dynamic texture learning with adaptive channel-discriminability for 3D mask face anti-spoofing. In 2017 IEEE International Joint Conference on Biometrics (IJCB) (pp. 748-755). IEEE.[16] Bak, C., Erdem, A., & Erdem, E. (2016). Two-stream convolutional networks for dynamic saliency prediction. arXiv preprint arXiv:1607.04730, 2(3), 6.PublicationORIGINALMethod for classifying images in databases through deep convolutional networks.pdfMethod for classifying images in databases through deep convolutional networks.pdfapplication/pdf1578152https://repositorio.cuc.edu.co/bitstreams/807d7e40-9468-47a3-a560-0e04ecdc534d/download952555b8fcb24fd3faac787b9ed11e3aMD51CC-LICENSElicense_rdflicense_rdfapplication/rdf+xml; charset=utf-8701https://repositorio.cuc.edu.co/bitstreams/4754bdec-6918-4c74-8c3d-4ecb85c9eee7/download42fd4ad1e89814f5e4a476b409eb708cMD52LICENSElicense.txtlicense.txttext/plain; charset=utf-83196https://repositorio.cuc.edu.co/bitstreams/846debed-f6c2-46c3-8e90-22f2b86c5afd/downloade30e9215131d99561d40d6b0abbe9badMD53THUMBNAILMethod for classifying images in databases through deep convolutional networks.pdf.jpgMethod for classifying images in databases through deep convolutional networks.pdf.jpgimage/jpeg45386https://repositorio.cuc.edu.co/bitstreams/9cef13a0-6159-45a0-bf5b-056c25f2915f/downloadb21703a4d16891344ec3fbb0751fd591MD54TEXTMethod for classifying images in databases through deep convolutional networks.pdf.txtMethod for classifying images in databases through deep convolutional networks.pdf.txttext/plain22510https://repositorio.cuc.edu.co/bitstreams/8f0da30c-d362-487c-bdee-a426422af3ad/downloadb29a11585eb13b8743c87c0719ab8f47MD5511323/7660oai:repositorio.cuc.edu.co:11323/76602024-09-16 16:45:41.44http://creativecommons.org/publicdomain/zero/1.0/CC0 1.0 Universalopen.accesshttps://repositorio.cuc.edu.coRepositorio de la Universidad de la Costa CUCrepdigital@cuc.edu.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 |