Coffee maturity classification using convolutional neural networks and transfer learning
This work presents a framework for coffee maturity classification from multispectral image data based on Convolutional Neural Networks (CNNs). The system leverages the use of multispectral image acquisition systems that generate large amounts of data, by taking advantage of the ability of CNNs to ex...
- Autores:
-
Tamayo Monsalve, Manuel Alejandro
Mercado Ruiz, Esteban
Villa Pulgarin, Juan Pablo
Bravo Ortíz, Mario Alejandro
Arteaga Arteaga, Harold Brayan
Mora Rubio, Alejandro
Alzate Grisales, Jesús Alejandro
Arias Garzón, Daniel
Romero Cano, Víctor
Orozco Arias, Simón
Osorio, Gustavo
Tabares Soto, Reinel
- Tipo de recurso:
- Article of journal
- Fecha de publicación:
- 2022
- Institución:
- Universidad Autónoma de Occidente
- Repositorio:
- RED: Repositorio Educativo Digital UAO
- Idioma:
- eng
- OAI Identifier:
- oai:red.uao.edu.co:10614/14730
- Acceso en línea:
- https://hdl.handle.net/10614/14730
https://red.uao.edu.co/
- Palabra clave:
- Redes neuronales (Computadores)
Neural networks (Computer science)
Coffee maturity classification
Convolutional neural network
Data augmentation
Deep learning
Multispectral images
Transfer learning
- Rights
- openAccess
- License
- Derechos reservados - IEEE, 2022
| id |
REPOUAO2_4b4031bdd2a10f0540faea0d036d94a3 |
|---|---|
| oai_identifier_str |
oai:red.uao.edu.co:10614/14730 |
| network_acronym_str |
REPOUAO2 |
| network_name_str |
RED: Repositorio Educativo Digital UAO |
| repository_id_str |
|
| dc.title.eng.fl_str_mv |
Coffee maturity classification using convolutional neural networks and transfer learning |
| title |
Coffee maturity classification using convolutional neural networks and transfer learning |
| spellingShingle |
Coffee maturity classification using convolutional neural networks and transfer learning Redes neuronales (Computadores) Neural networks (Computer science) Coffee maturity classification Convolutional neural network Data augmentation Deep learning Multispectral images Transfer learning |
| title_short |
Coffee maturity classification using convolutional neural networks and transfer learning |
| title_full |
Coffee maturity classification using convolutional neural networks and transfer learning |
| title_fullStr |
Coffee maturity classification using convolutional neural networks and transfer learning |
| title_full_unstemmed |
Coffee maturity classification using convolutional neural networks and transfer learning |
| title_sort |
Coffee maturity classification using convolutional neural networks and transfer learning |
| dc.creator.fl_str_mv |
Tamayo Monsalve, Manuel Alejandro Mercado Ruiz, Esteban Villa Pulgarin, Juan Pablo Bravo Ortíz, Mario Alejandro Arteaga Arteaga, Harold Brayan Mora Rubio, Alejandro Alzate Grisales, Jesús Alejandro Arias Garzón, Daniel Romero Cano, Víctor Orozco Arias, Simón Osorio, Gustavo Tabares Soto, Reinel |
| dc.contributor.author.none.fl_str_mv |
Tamayo Monsalve, Manuel Alejandro Mercado Ruiz, Esteban Villa Pulgarin, Juan Pablo Bravo Ortíz, Mario Alejandro Arteaga Arteaga, Harold Brayan Mora Rubio, Alejandro Alzate Grisales, Jesús Alejandro Arias Garzón, Daniel Romero Cano, Víctor Orozco Arias, Simón Osorio, Gustavo Tabares Soto, Reinel |
| dc.subject.armarc.spa.fl_str_mv |
Redes neuronales (Computadores) |
| topic |
Redes neuronales (Computadores) Neural networks (Computer science) Coffee maturity classification Convolutional neural network Data augmentation Deep learning Multispectral images Transfer learning |
| dc.subject.armarc.eng.fl_str_mv |
Neural networks (Computer science) |
| dc.subject.proposal.eng.fl_str_mv |
Coffee maturity classification Convolutional neural network Data augmentation Deep learning Multispectral images Transfer learning |
| description |
This work presents a framework for coffee maturity classification from multispectral image data based on Convolutional Neural Networks (CNNs). The system leverages the use of multispectral image acquisition systems that generate large amounts of data, by taking advantage of the ability of CNNs to extract meaningful patterns from very high-dimensional data. We validated the use of five different popular CNN architectures on the classification of cherry coffee fruits according to their ripening stage. The different models were trained on a training dataset balanced in different ways, which resulted in a top accuracy higher than 98% when applied to the classification of 600 coffee fruits in 5 different stages of ripening. This work has the potential of providing the farmer with a high-quality, optimized, accurate and viable method for classifying coffee fruits. In order to foster future research in this area, the data used in this work, which was acquired with a custom-developed multispectral image acquisition system, have been released |
| publishDate |
2022 |
| dc.date.issued.none.fl_str_mv |
2022-04 |
| dc.date.accessioned.none.fl_str_mv |
2023-05-11T18:51:22Z |
| dc.date.available.none.fl_str_mv |
2023-05-11T18:51:22Z |
| 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.coarversion.fl_str_mv |
http://purl.org/coar/version/c_970fb48d4fbd8a85 |
| 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/publishedVersion |
| format |
http://purl.org/coar/resource_type/c_6501 |
| status_str |
publishedVersion |
| dc.identifier.issn.spa.fl_str_mv |
21693536 |
| dc.identifier.uri.none.fl_str_mv |
https://hdl.handle.net/10614/14730 |
| dc.identifier.instname.spa.fl_str_mv |
Universidad Autónoma de Occidente |
| dc.identifier.reponame.spa.fl_str_mv |
Repositorio Educativo Digital UAO |
| dc.identifier.repourl.spa.fl_str_mv |
https://red.uao.edu.co/ |
| identifier_str_mv |
21693536 Universidad Autónoma de Occidente Repositorio Educativo Digital UAO |
| url |
https://hdl.handle.net/10614/14730 https://red.uao.edu.co/ |
| dc.language.iso.spa.fl_str_mv |
eng |
| language |
eng |
| dc.relation.citationendpage.spa.fl_str_mv |
42982 |
| dc.relation.citationstartpage.spa.fl_str_mv |
42971 |
| dc.relation.citationvolume.spa.fl_str_mv |
10 |
| dc.relation.cites.none.fl_str_mv |
Tamayo Monsalve, M. A., Mercado Ruiz, E., Villa Pulgarin, J. P., Bravo Ortíz., M. A., Arteaga, H. B. Arteaga. A. Mora Rubio. Alzate Grisales, J. A., Arias Garzon., D. Romero Cano, V., Orozco Arias, S., Osorio, G., Tabares Soto, R. (2022). Coffee Maturity Classification Using Convolutional Neural Networks and Transfer Learning. IEEE Access, 10, 42971-42982. https://hdl.handle.net/10614/14730 |
| dc.relation.ispartofjournal.spa.fl_str_mv |
IEEE Access |
| dc.relation.references.spa.fl_str_mv |
G. Gyarmati and T. Mizik, ‘‘The present and future of the precision agri- culture,’’ in Proc. IEEE 15th Int. Conf. Syst. Syst. Eng. (SoSE), Jun. 2020, pp. 593–596 S. Cubero, N. Aleixos, E. Moltó, J. Gómez-Sanchis, and J. Blasco, ‘‘Advances in machine vision applications for automatic inspection and quality evaluation of fruits and vegetables,’’ Food Bioprocess Technol., vol. 4, no. 4, pp. 487–504, May 2011. Y. A. Ohali, ‘‘Computer vision based date fruit grading system: Design and implementation,’’ J. King Saud Univ., Comput. Inf. Sci., vol. 23, no. 1, pp. 29–36, Jan. 2011 D. Wu and D.-W. Sun, ‘‘Advanced applications of hyperspectral imag- ing technology for food quality and safety analysis and assessment: A review—Part I: Fundamentals,’’ Innov. Food Sci. Emerg. Technol., vol. 19, pp. 1–14, Jul. 2013. L. B. Furstenau, M. K. Sott, L. M. Kipper, E. L. Machado, J. R. Lopez-Robles, M. S. Dohan, M. J. Cobo, A. Zahid, Q. H. Abbasi, and M. A. Imran, ‘‘Link between sustainability and industry 4.0: Trends, challenges and new perspectives,’’ IEEE Access, vol. 8, pp. 140079–140096, 2020. S. Munera, C. Besada, J. Blasco, S. Cubero, A. Salvador, P. Talens, and N. Aleixos, ‘‘Astringency assessment of persimmon by hyperspectral imaging,’’ Postharvest Biol. Technol., vol. 125, pp. 35–41, Mar. 2017. M. Taghizadeh, A. A. Gowen, and C. P. O’Donnell, ‘‘Comparison of hyperspectral imaging with conventional RGB imaging for quality eval- uation of Agaricus bisporus mushrooms,’’ Biosyst. Eng., vol. 108, no. 2, pp. 191–194, Feb. 2011 S. Ponte, ‘‘Estándares, comercio y equidad: Lecciones de la industria de los cafés especiales,’’ Economía Mundial del café, Centro de Investigaciones Para el Desarrollo de Copenhague, Anaheim, CF, Tech. Rep. 5 de mayo de, 2002, pp. 131–163 M. Sott, L. Furstenau, L. Kipper, F. Giraldo, J. Lpez-Robles, M. Cobo, A. Zahid, Q. Abbasi, and M. Imran, ‘‘Precision techniques and agri- culture 4.0 technologies to promote sustainability in the coffee sector: State of the art, challenges and future trends,’’ IEEE Access, vol. 8, pp. 149854–149867, 2020. A. G. Costa, D. A. G. D. Sousa, J. L. Paes, J. P. B. Cunha, and M. V. M. D. Oliveira, ‘‘Classification of robusta coffee fruits at different maturation stages using colorimetric characteristics,’’ Engenharia Agrí- cola, vol. 40, no. 4, pp. 518–525, Aug. 2020 L. Cavigelli, D. Bernath, M. Magno, and L. Benini, ‘‘Computationally efficient target classification in multispectral image data with deep neural networks,’’ CoRR, vol. 10, 2016 A. H. Shahin, A. Kamal, and M. A. Elattar, ‘‘Deep ensemble learning for skin lesion classification from dermoscopic images,’’ in Proc. 9th Cairo Int. Biomed. Eng. Conf. (CIBEC), Dec. 2018, pp. 150–153 J. P. Rodríguez, D. C. Corrales, J.-N. Aubertot, and J. C. Corrales, ‘‘A com- puter vision system for automatic cherry beans detection on coffee trees,’’ Pattern Recognit. Lett., vol. 136, pp. 142–153, Aug. 2020. Z. Huo, G. Du, F. Luo, Y. Qiao, and J. Luo, ‘‘D-MSCD: Mean-standard deviation curve descriptor based on deep learning,’’ IEEE Access, vol. 8, pp. 204509–204517, 2020 |
| dc.rights.spa.fl_str_mv |
Derechos reservados - IEEE, 2022 |
| dc.rights.coar.fl_str_mv |
http://purl.org/coar/access_right/c_abf2 |
| dc.rights.uri.spa.fl_str_mv |
https://creativecommons.org/licenses/by-nc-nd/4.0/ |
| dc.rights.accessrights.spa.fl_str_mv |
info:eu-repo/semantics/openAccess |
| dc.rights.creativecommons.spa.fl_str_mv |
Atribución-NoComercial-SinDerivadas 4.0 Internacional (CC BY-NC-ND 4.0) |
| rights_invalid_str_mv |
Derechos reservados - IEEE, 2022 https://creativecommons.org/licenses/by-nc-nd/4.0/ Atribución-NoComercial-SinDerivadas 4.0 Internacional (CC BY-NC-ND 4.0) http://purl.org/coar/access_right/c_abf2 |
| eu_rights_str_mv |
openAccess |
| dc.format.extent.spa.fl_str_mv |
12 páginas |
| dc.format.mimetype.spa.fl_str_mv |
application/pdf |
| dc.publisher.spa.fl_str_mv |
IEEE |
| institution |
Universidad Autónoma de Occidente |
| bitstream.url.fl_str_mv |
https://red.uao.edu.co/bitstreams/f3e7b50f-927e-47a2-8c13-ab000b9f168f/download https://red.uao.edu.co/bitstreams/0b8bf46c-0f22-4b2c-802d-0c0486de9fe0/download https://red.uao.edu.co/bitstreams/b36af18e-6d68-4b48-aa9a-f6fa4a65d538/download https://red.uao.edu.co/bitstreams/97c97f22-bd95-4695-a2eb-16227e51a4c0/download |
| bitstream.checksum.fl_str_mv |
93808f2db1fc820458411614c0a20e62 20b5ba22b1117f71589c7318baa2c560 272d0d55ac8c85d6468953ca51e8485a e6acf4637f6769dc9d4741b62e07995b |
| bitstream.checksumAlgorithm.fl_str_mv |
MD5 MD5 MD5 MD5 |
| repository.name.fl_str_mv |
Repositorio Digital Universidad Autonoma de Occidente |
| repository.mail.fl_str_mv |
repositorio@uao.edu.co |
| _version_ |
1814260017878007808 |
| spelling |
Tamayo Monsalve, Manuel Alejandro6b905e9c60929c6fee3ea5a25474f77bMercado Ruiz, Esteban79a149e7e68595ee144cc0eb66897184Villa Pulgarin, Juan Pablo71fcb77691fb889a1e38604be5293a5aBravo Ortíz, Mario Alejandrofa72add33fa480a7780d9a9dad24e5e4Arteaga Arteaga, Harold Brayan4a94f99f494e44119ad843f4f82d1fd5Mora Rubio, Alejandro24d14ce7d99f40e5774328bf81af3821Alzate Grisales, Jesús Alejandro39e5efd4131c23f57386bc2e978d3d0bArias Garzón, Daniel8fe49ba9e2899b17037e0f9fa8f0687fRomero Cano, Víctorac8a4e8955699e0474b9d0266969e148Orozco Arias, SimónOsorio, Gustavoff632fa9f70ee8fad6f38268a81bb956Tabares Soto, Reinelab800effcb910eccdd754c6d1ed7b2472023-05-11T18:51:22Z2023-05-11T18:51:22Z2022-0421693536https://hdl.handle.net/10614/14730Universidad Autónoma de OccidenteRepositorio Educativo Digital UAOhttps://red.uao.edu.co/This work presents a framework for coffee maturity classification from multispectral image data based on Convolutional Neural Networks (CNNs). The system leverages the use of multispectral image acquisition systems that generate large amounts of data, by taking advantage of the ability of CNNs to extract meaningful patterns from very high-dimensional data. We validated the use of five different popular CNN architectures on the classification of cherry coffee fruits according to their ripening stage. The different models were trained on a training dataset balanced in different ways, which resulted in a top accuracy higher than 98% when applied to the classification of 600 coffee fruits in 5 different stages of ripening. This work has the potential of providing the farmer with a high-quality, optimized, accurate and viable method for classifying coffee fruits. In order to foster future research in this area, the data used in this work, which was acquired with a custom-developed multispectral image acquisition system, have been released 12 páginasapplication/pdfengIEEEDerechos reservados - IEEE, 2022https://creativecommons.org/licenses/by-nc-nd/4.0/info:eu-repo/semantics/openAccessAtribución-NoComercial-SinDerivadas 4.0 Internacional (CC BY-NC-ND 4.0)http://purl.org/coar/access_right/c_abf2Coffee maturity classification using convolutional neural networks and transfer learningArtículo 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/publishedVersionhttp://purl.org/coar/version/c_970fb48d4fbd8a85Redes neuronales (Computadores)Neural networks (Computer science)Coffee maturity classificationConvolutional neural networkData augmentationDeep learningMultispectral imagesTransfer learning429824297110Tamayo Monsalve, M. A., Mercado Ruiz, E., Villa Pulgarin, J. P., Bravo Ortíz., M. A., Arteaga, H. B. Arteaga. A. Mora Rubio. Alzate Grisales, J. A., Arias Garzon., D. Romero Cano, V., Orozco Arias, S., Osorio, G., Tabares Soto, R. (2022). Coffee Maturity Classification Using Convolutional Neural Networks and Transfer Learning. IEEE Access, 10, 42971-42982. https://hdl.handle.net/10614/14730IEEE AccessG. Gyarmati and T. Mizik, ‘‘The present and future of the precision agri- culture,’’ in Proc. IEEE 15th Int. Conf. Syst. Syst. Eng. (SoSE), Jun. 2020, pp. 593–596S. Cubero, N. Aleixos, E. Moltó, J. Gómez-Sanchis, and J. Blasco, ‘‘Advances in machine vision applications for automatic inspection and quality evaluation of fruits and vegetables,’’ Food Bioprocess Technol., vol. 4, no. 4, pp. 487–504, May 2011.Y. A. Ohali, ‘‘Computer vision based date fruit grading system: Design and implementation,’’ J. King Saud Univ., Comput. Inf. Sci., vol. 23, no. 1, pp. 29–36, Jan. 2011D. Wu and D.-W. Sun, ‘‘Advanced applications of hyperspectral imag- ing technology for food quality and safety analysis and assessment: A review—Part I: Fundamentals,’’ Innov. Food Sci. Emerg. Technol., vol. 19, pp. 1–14, Jul. 2013.L. B. Furstenau, M. K. Sott, L. M. Kipper, E. L. Machado, J. R. Lopez-Robles, M. S. Dohan, M. J. Cobo, A. Zahid, Q. H. Abbasi, and M. A. Imran, ‘‘Link between sustainability and industry 4.0: Trends, challenges and new perspectives,’’ IEEE Access, vol. 8, pp. 140079–140096, 2020.S. Munera, C. Besada, J. Blasco, S. Cubero, A. Salvador, P. Talens, and N. Aleixos, ‘‘Astringency assessment of persimmon by hyperspectral imaging,’’ Postharvest Biol. Technol., vol. 125, pp. 35–41, Mar. 2017.M. Taghizadeh, A. A. Gowen, and C. P. O’Donnell, ‘‘Comparison of hyperspectral imaging with conventional RGB imaging for quality eval- uation of Agaricus bisporus mushrooms,’’ Biosyst. Eng., vol. 108, no. 2, pp. 191–194, Feb. 2011S. Ponte, ‘‘Estándares, comercio y equidad: Lecciones de la industria de los cafés especiales,’’ Economía Mundial del café, Centro de Investigaciones Para el Desarrollo de Copenhague, Anaheim, CF, Tech. Rep. 5 de mayo de, 2002, pp. 131–163M. Sott, L. Furstenau, L. Kipper, F. Giraldo, J. Lpez-Robles, M. Cobo, A. Zahid, Q. Abbasi, and M. Imran, ‘‘Precision techniques and agri- culture 4.0 technologies to promote sustainability in the coffee sector: State of the art, challenges and future trends,’’ IEEE Access, vol. 8, pp. 149854–149867, 2020.A. G. Costa, D. A. G. D. Sousa, J. L. Paes, J. P. B. Cunha, and M. V. M. D. Oliveira, ‘‘Classification of robusta coffee fruits at different maturation stages using colorimetric characteristics,’’ Engenharia Agrí- cola, vol. 40, no. 4, pp. 518–525, Aug. 2020L. Cavigelli, D. Bernath, M. Magno, and L. Benini, ‘‘Computationally efficient target classification in multispectral image data with deep neural networks,’’ CoRR, vol. 10, 2016A. H. Shahin, A. Kamal, and M. A. Elattar, ‘‘Deep ensemble learning for skin lesion classification from dermoscopic images,’’ in Proc. 9th Cairo Int. Biomed. Eng. Conf. (CIBEC), Dec. 2018, pp. 150–153J. P. Rodríguez, D. C. Corrales, J.-N. Aubertot, and J. C. Corrales, ‘‘A com- puter vision system for automatic cherry beans detection on coffee trees,’’ Pattern Recognit. Lett., vol. 136, pp. 142–153, Aug. 2020.Z. Huo, G. Du, F. Luo, Y. Qiao, and J. Luo, ‘‘D-MSCD: Mean-standard deviation curve descriptor based on deep learning,’’ IEEE Access, vol. 8, pp. 204509–204517, 2020Comunidad generalPublicationORIGINALCoffee_Maturity_Classification_Using_Convolutional_Neural_Networks_and_Transfer_Learning.pdfCoffee_Maturity_Classification_Using_Convolutional_Neural_Networks_and_Transfer_Learning.pdftexto completo del artículoapplication/pdf2647218https://red.uao.edu.co/bitstreams/f3e7b50f-927e-47a2-8c13-ab000b9f168f/download93808f2db1fc820458411614c0a20e62MD51LICENSElicense.txtlicense.txttext/plain; charset=utf-81665https://red.uao.edu.co/bitstreams/0b8bf46c-0f22-4b2c-802d-0c0486de9fe0/download20b5ba22b1117f71589c7318baa2c560MD52TEXTCoffee_Maturity_Classification_Using_Convolutional_Neural_Networks_and_Transfer_Learning.pdf.txtCoffee_Maturity_Classification_Using_Convolutional_Neural_Networks_and_Transfer_Learning.pdf.txtExtracted texttext/plain61245https://red.uao.edu.co/bitstreams/b36af18e-6d68-4b48-aa9a-f6fa4a65d538/download272d0d55ac8c85d6468953ca51e8485aMD53THUMBNAILCoffee_Maturity_Classification_Using_Convolutional_Neural_Networks_and_Transfer_Learning.pdf.jpgCoffee_Maturity_Classification_Using_Convolutional_Neural_Networks_and_Transfer_Learning.pdf.jpgGenerated Thumbnailimage/jpeg15906https://red.uao.edu.co/bitstreams/97c97f22-bd95-4695-a2eb-16227e51a4c0/downloade6acf4637f6769dc9d4741b62e07995bMD5410614/14730oai:red.uao.edu.co:10614/147302024-04-02 16:45:12.833https://creativecommons.org/licenses/by-nc-nd/4.0/Derechos reservados - IEEE, 2022open.accesshttps://red.uao.edu.coRepositorio Digital Universidad Autonoma de Occidenterepositorio@uao.edu.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 |