Las Redes neuronales convolucionales ResNet-50 para la detección de gorgojo en granos de maíz

Iván Alberto Analuisa Aroca; Arnaldo  Vergara-Romero; Iris Betzaida  Pérez Almeida

doi:10.17268/sci.agropecu.2023.034

Autores

Iván Alberto Analuisa Aroca Universidad ECOTEC. https://orcid.org/0000-0002-3798-3122
Arnaldo Vergara-Romero Universidad ECOTEC
Iris Betzaida Pérez Almeida Universidad ECOTEC

DOI:

https://doi.org/10.17268/sci.agropecu.2023.034

Palavras-chave:

Gorgulho, milho, redes neurais convolucionais, Equador.

Resumo

No campo da informação agrícola, a conservação e o diagnóstico precoce de doenças dos grãos de milho são desejáveis. As causas de danos por agentes externos são um problema no setor agrícola. Na detecção de pragas e redução dos efeitos sobre os grãos, o aprendizado profundo dentro da inteligência artificial (IA) é usado no controle de qualidade dos grãos, ajudando a fazer análises de produção para a tomada de decisões. As imagens são utilizadas para classificar diferentes grãos de milho, identificando aqueles danificados por gorgulhos ou outras pragas. Neste documento, um modelo de rede convolucional é proposto com base na aprendizagem do reconhecimento de padrões na presença de grãos associados a danos do gorgulho no grão. Resultados satisfatórios são obtidos com taxas de precisão de 100% (amostra de treinamento), 97% (amostra de validação) e 98% (conjunto de testes). A precisão, sensibilidade, especificidade, índice de qualidade, AUC e F-score da ResNet-50 foram 0,9464, 0,9310, 0,9630, 0,9469, 0,9470 e 0,9474 respectivamente. As principais conclusões mostram que os parâmetros do modelo melhorado são significativos, o reconhecimento do gorgulho em grãos com o modelo tem uma precisão de identificação significativa. Os agentes econômicos da cadeia de valor dão mais importância às relações comerciais com clientes e fornecedores do que à qualidade e preservação dos grãos atualmente importantes para a competitividade.

Referências

Abiyev, R. H., & Maaitah, M. K. S. (2018). Deep Convolutional Neural Networks for Chest Diseases Detection. Journal of Healthcare Engineering, 2018. doi:10.1155/2018/4168538.

Aladhadh, S., Habib, S., Islam, M., Aloraini, M., Aladhadh, M., & Al-Rawashdeh, H. S. (2022). An Efficient Pest Detection Framework with a Medium-Scale Benchmark to Increase the Agricultural Productivity. Sensors, 22(24), 9749. doi:10.3390/s22249749.

Alchetron (2022). Maize weevil. https://alchetron.com/Maize-weevil.

Altuntaş, Y., Cömert, Z., & Kocamaz, A. F. (2019). Identification of haploid and diploid maize seeds using convolutional neural networks and a transfer learning approach. Computers and Electronics in Agriculture, 163, 104874. doi:10.1016/j.compag.2019.104874.

An, J., Li, W., Li, M., Cui, S., & Yue, H. (2019). Identification and classification of maize drought stress using deep convolutional neural network. Symmetry, 11(2), 256. doi:10.3390/sym11020256.

Baoua, I. B., Amadou, L., Ousmane, B., Baributsa, D., & Murdock, L. L. (2014). PICS bags for post-harvest storage of maize grain in West Africa. Journal of Stored Products Research, 58, 20–28. doi:10.1016/j.jspr.2014.03.001.

Basir, M. S., Chowdhury, M., Islam, M. D., & Ashik-E-Rabbani, M. (2021). Artificial neural network model in predicting yield of mechanically transplanted rice from transplanting parameters in Bangladesh. Journal of Agriculture and Food Research, 5, 100186. doi:10.1016/j.jafr.2021.100186.

Bisong, E. (2019). Building Machine Learning and Deep Learning Models on Google Cloud Platform. 1st edn, Building Machine Learning and Deep Learning Models on Google Cloud Platform. 1st edn. Ottawa: Apress. doi:10.1007/978-1-4842-4470-8.

Bohinc, T., Horvat, A., Andrić, G., Pražić Golić, M., Kljajić, P., & Trdan, S. (2020). Natural versus synthetic zeolites for controlling the maize weevil (Sitophilus zeamais)–like Messi versus Ronaldo? Journal of Stored Products Research, 88, 101639.

Camardo, M., Mazzoni, M., & Battilani, P. (2021). Machine Learning for Predicting Mycotoxin Occurrence in Maize. Frontiers in Microbiology, 12, 1–10. doi:10.3389/fmicb.2021.661132.

Coulibaly, S., Kamsu-Foguem, B., Kamissoko, D., Traore, D. (2022). Explainable deep convolutional neural networks for insect pest recognition. Journal of Cleaner Production, 371, 133638. doi:10.1016/j.jclepro.2022.133638.

da Silva, C. B., Silva, A. A. N., Barroso, G., Yamamoto, P. T., Arthur, V., Toledo, C. F. M., & Mastrangelo, T. de A. (2021). Convolutional neural networks using enhanced radiographs for real-time detection of Sitophilus zeamais in maize grain. Foods, 10(4), 879. doi:10.3390/foods10040879.

de Carvalho, M. L., Rezende Leite, E., Carvalho, G. A., França-Silva, F., Bernardes de Andrade, D., & Marques, E. R. (2019). The Compared Efficiency of the Traditional Method, Radiography without Contrast and Radiography with Contrast in the Determination of Infestation by Weevil (Sitophilus zeamais) in Maize Seeds. Insects, 10(6), 156.

FAO (2017). El futuro de la alimentación y la agricultura: Tendencias y desafíos. Organización de las Naciones Unidas para la Alimentación y la Agricultura. https://www.fao.org/3/i6881s/i6881s.pdf

Fan, J., Zheng, J., Wu, L., & Zhang, F. (2021). Estimation of daily maize transpiration using support vector machines, extreme gradient boosting, artificial and deep neural networks models. Agricultural Water Management, 245, 106547. doi:10.1016/j.agwat.2020.106547.

Goodfellow, I., Bengio, Y., & Courville, A. (2017) Deep Learning, The MIT Press. Londres: MIT Press. doi:10.1017/CBO9781107415324.004.

Ileke, K. D., Idoko, J. E., Ojo, D. O., & Adesina, B. C. (2020). Evaluation of botanical powders and extracts from Nigerian plants as protectants of maize grains against maize weevil, Sitophilus zeamais (Motschulsky) [Coleoptera: Curculionidae]. Biocatal. Agric. Biotechnol., 27, 101702.

Javanmardi, S., Miraei Ashtiani, S.-H., Verbeek, F. J., & Martynenko, A. (2021). Computer-vision classification of corn seed varieties using deep convolutional neural network’, Journal of Stored Products Research., 92, 101800. doi:10.1016/j.jspr.2021.101800.

Kamilaris, A., & Prenafeta, F. X. (2018). A review of the use of convolutional neural networks in agriculture. Journal of Agricultural Science, 156(3), 312–322. doi:10.1017/S0021859618000436.

Kienbaum, L., Correa Abondano, M., Blas, R., & Schmid, K. (2021). DeepCob: precise and high-throughput analysis of maize cob geometry using deep learning with an application in genebank phenomics. Plant Methods, 17, 91. doi:10.1186/s13007-021-00787-6.

Li, L., Lv, Ch., Yuan, Y., & Zhao, B. (2021). Maize residue segmentation using Siamese domain transfer network. Computers and Electronics in Agriculture, 187, 106261. doi:10.1016/j.compag.2021.106261.

Mamoon-ur-Rashid, M., Riaz-ud-din, Tariq, M., Khan, A. A., Latif, A., Naeem, M., & Khan, I. (2021). Bioactivity of hexane plant extracts against maize weevil (Sitophilus zeamais Motschulsky) (Coleoptera: Curculionidae) on stored maize. Intl J Agric Biol, 26, 617‒624.

Mi, C., Zhao, Ch., Deng, Q., & Deng, X. (2021). Prediction of chilling damage risk in maize growth period based on probabilistic neural network approach. International Journal of Agricultural and Biological Engineering, 14(2), 120–125. doi:10.25165/J.IJABE.20211402.5732.

Michelucci, U. (2018). Applied Deep Learning: A Case-Based Approach to Understanding Deep Neural Networks. Dübendorf, Switzerland. doi:https://doi.org/10.1007/978-1-4842-3790-8.

Ministerio de Agricultura y Ganadería, (MAG) (2019). Ficha del cultivo de Maíz duro seco (Zea mays L.). Quito-Ecuador, p. 1. http://sipa.agricultura.gob.ec/index.php/cifras-agroproductivas.

Muhaisin, M. M., & Rahman, T. (2019). Application of Game Theory for Big Data Analytics, Data Analytics. doi:10.1201/9780429446177-8.

Ni, C., Wang, D., Vinson, R., Holmes, M., & Tao, Y. (2019). Automatic inspection machine for maize kernels based on deep convolutional neural networks. Biosystems Engineering, 178, 131–144. doi:10.1016/j.biosystemseng.2018.11.010.

Ngom, D., Fauconnier, M. L., Malumba, P., Thiaw, C., Brévault, T., & Sembène, M. (2021). Morphophysical and biochemical traits involved in maize grain varietal susceptibility to the maize weevil, Sitophilus zeamais (Coleoptera, Curculionidae). Biotechnologie, Agronomie, Société et Environnement, 25(1), 45-56.

Nurma Yulita, I., Hidayat, A., Setiawan Abdullah, A., & Paulus, E. (2020). Combining Fuzzy Clustering and Hidden Markov Models for Sundanese Speech Recognition. Journal of Physics: Conference Series, 1028, 012239.

Nyéki, A., Kerepesi, C., Daróczy, B., Benczúr, A., Milics, G., Nagy, J., Harsányi, E., Kovács, A. J., & Neményi, M. (2021). Application of spatio-temporal data in site-specific maize yield prediction with machine learning methods. Precision Agriculture, 22(5), 1397–1415. doi:10.1007/s11119-021-09833-8.

Pashaei, M., Kamangir, H., Starek, M. J., & Tissot, P. (2020). Review and evaluation of deep learning architectures for efficient land cover mapping with UAS hyper-spatial imagery: A case study over a wetland. Remote Sensing, 12(6), 959. doi:10.3390/rs12060959.

Peschiutta, M. L., Brito, V. D., Achimón, F., Zunino, M. P., Usseglio, V. L., & Zygadlo, J. A. (2019). New insecticide delivery method for the control of Sitophilus zeamais in stored maize. Journal of Stored Products Research, 83, 185-190.

Rau, T. S., Bern, C. J., Brumm, T. J., Barnes, R. B., Bbosa, D., & Maier, D. E. (2021). Evaluation of stirring to suppress weevils in stored maize. Journal of Stored Products Research, 93, 101849.

Rimal, K., Shah, K. B., & Jha, A. K. (2023). Advanced multi-class deep learning convolution neural network approach for insect pest classification using TensorFlow. International Journal of Environmental Science and Technology, 20(4), 4003–4016. doi:10.1007/s13762-022-04277-7.

Sibiya, M., & Sumbwanyambe, M. (2019). A Computational Procedure for the Recognition and Classification of Maize Leaf Diseases Out of Healthy Leaves Using Convolutional Neural Networks. AgriEngineering, 1(1), 119–131. doi:10.3390/agriengineering1010009.

Sibiya, M., & Sumbwanyambe, M. (2021). Automatic fuzzy logic-based maize common rust disease severity predictions with thresholding and deep learning. Pathogens, 10(2), 131. doi:10.3390/pathogens10020131.

Stathers, T. E., Arnold, S. E. J., Rumney, C. J., & Hopson, C., (2020). Measuring the nutritional cost of insect infestation of stored maize and cowpea, Food Security: The Science, Sociology and Economics of Food Production and Access to Food, 12(2), 285-308.

Stuhl, Ch. J. (2019). Does prior feeding behavior by previous generations of the maize weevil (Coleoptera: Curculionidae) determine future descendants feeding preference and ovipositional suitability? Florida Entomologist, 102(2), 366-372.

Srivastava, N., Hinton, G., Krizhevsky, A., Sutskever, I., Salakhutdinov, R. (2014). Dropout: A simple way to prevent neural networks from overfitting. Journal of Machine Learning Research, 15, 1929-1958.

Waheed, A., Goyal, M., Gupta, D., Khanna, A., Ella Hassanien, A., & Mohan Pandey, H. (2020). An optimized dense convolutional neural network model for disease recognition and classification in corn leaf. Computers and Electronics in Agriculture, 175, 105456. doi:10.1016/j.compag.2020.105456.

Wang, G., Sun, Y., & Wang, J. (2017). Automatic Image-Based Plant Disease Severity Estimation Using Deep Learning. Computational Intelligence and Neuroscience, 2917536. doi:10.1155/2017/2917536.

White, T. (2018) Hadoop. The definitive guide. Storage and analysis at internet scale. 4th Editio. Edited by Oreilly. Sebastopol-USA. http://oreilly.com/catalog/errata.csp?isbn=9781491901632.

Xia, D., Chen, P., Wang, B., Zhang, J., & Xie, Ch. (2018). Insect detection and classification based on an improved convolutional neural network. Sensors (Switzerland), 18(12), 4169. doi:10.3390/s18124169.

Xu, Y., Zhao, B., Zhai, Y., Chen, Q., & Zhou, Y. (2021). Maize Diseases Identification Method Based on Multi-Scale Convolutional Global Pooling Neural Network. IEEE Access, 9, 27959–27970. doi:10.1109/ACCESS.2021.3058267.

Yang, W., Nigon, T., Hao, Z., Dias Paiao, G., Fernández, F. G., Mulla, D., & Yang, C. (2021). Estimation of corn yield based on hyperspectral imagery and convolutional neural network. Computers and Electronics in Agriculture, 184, 106092 doi:10.1016/j.compag.2021.106092.

Zhang, J., Ma, Q., Cui, X., Guo, H., Wang, K., & Zhu, D. (2020). High-throughput corn ear screening method based on two-pathway convolutional neural network. Computers and Electronics in Agriculture, 175, 105525. doi:10.1016/j.compag.2020.105525.

Zhang, X., Qiao, Y., Meng, F., Fan, C., & Zhang, M. (2018). Identification of maize leaf diseases using improved deep convolutional neural networks. IEEE Access, 6, 30370–30377. doi:10.1109/ACCESS.2018.2844405.