Oat grains classification using deep learning

IF 0.2 Q4 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS Revista Brasileira de Computacao Aplicada Pub Date : 2023-04-25 DOI:10.5335/rbca.v15i1.13653
D. I. Patrício, Carlos Ré Signor, N. C. Lângaro, Rafael Rieder
{"title":"Oat grains classification using deep learning","authors":"D. I. Patrício, Carlos Ré Signor, N. C. Lângaro, Rafael Rieder","doi":"10.5335/rbca.v15i1.13653","DOIUrl":null,"url":null,"abstract":"Background: Based on their nutritional benefits, oat is classified as a cereal of great importance for both human and animal feeding. Throughout the production process, species and variety identification are vital for agricultural systems. The present work establishes SeedFlow, a method for image acquisition, processing, and classification of oat grains using deep learning techniques. We apply these techniques to the identification of the grains from the different oat species Avena sativa and Avena strigosa and to classify grains as varieties of Avena sativa, such as UPFA Ouro, UPFA Fuerza, and UPFA Gaudéria. Results: To achieve this proposition, we executed our solution considering six different deep learning architectures to evaluate which model presents the best performance. This approach attained an accuracy of 99.7% for oat species identification and 89.7% for oat varieties classification using DenseNet architecture. Conclusions: As a result, this tool can provide high value for practical quality control applications, and it is feasible to use in pre-screening tests, laboratory analysis pipelines, or computer support tools geared toward breeding programs or intellectual property assessment.","PeriodicalId":41711,"journal":{"name":"Revista Brasileira de Computacao Aplicada","volume":null,"pages":null},"PeriodicalIF":0.2000,"publicationDate":"2023-04-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Revista Brasileira de Computacao Aplicada","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.5335/rbca.v15i1.13653","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS","Score":null,"Total":0}
引用次数: 0

Abstract

Background: Based on their nutritional benefits, oat is classified as a cereal of great importance for both human and animal feeding. Throughout the production process, species and variety identification are vital for agricultural systems. The present work establishes SeedFlow, a method for image acquisition, processing, and classification of oat grains using deep learning techniques. We apply these techniques to the identification of the grains from the different oat species Avena sativa and Avena strigosa and to classify grains as varieties of Avena sativa, such as UPFA Ouro, UPFA Fuerza, and UPFA Gaudéria. Results: To achieve this proposition, we executed our solution considering six different deep learning architectures to evaluate which model presents the best performance. This approach attained an accuracy of 99.7% for oat species identification and 89.7% for oat varieties classification using DenseNet architecture. Conclusions: As a result, this tool can provide high value for practical quality control applications, and it is feasible to use in pre-screening tests, laboratory analysis pipelines, or computer support tools geared toward breeding programs or intellectual property assessment.
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
燕麦颗粒分类的深度学习方法
背景:基于其营养价值,燕麦被归类为对人类和动物喂养都非常重要的谷物。在整个生产过程中,物种和品种鉴定对农业系统至关重要。目前的工作建立了SeedFlow,一种使用深度学习技术对燕麦颗粒进行图像采集、处理和分类的方法。我们将这些技术应用于不同燕麦品种Avena sativa和Avena strigosa的籽粒鉴定,并将籽粒分类为Avena sativa的品种,如UPFA Ouro, UPFA Fuerza和UPFA gaud。结果:为了实现这一命题,我们考虑了六种不同的深度学习架构来执行我们的解决方案,以评估哪种模型表现出最佳性能。该方法对燕麦品种鉴定的准确率为99.7%,对DenseNet结构的燕麦品种分类准确率为89.7%。结论:该工具可为实际质量控制应用提供高价值,并且可用于预筛选测试、实验室分析管道或面向育种计划或知识产权评估的计算机支持工具。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 去求助
来源期刊
Revista Brasileira de Computacao Aplicada
Revista Brasileira de Computacao Aplicada COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS-
自引率
50.00%
发文量
18
期刊最新文献
GRSR - a guideline for reporting studies results for machine learning applied to Electroencephalogram data Detecção e alerta de equipamentos não permitidos em quartos hospitalares por meio da supervisão da corrente elétrica Otimização inspirada na interação ecológica de predação do gato em relação ao rato aplicada ao problema da múltipla mochila 0-1 Classificação de sinais de voz para auxílio no diagnóstico da doença de Parkinson Authorship attribution of comments in Portuguese extracted from Reddit
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1