用于烟雾和野火探测的 YOLOv8 超参数优化:对农业和环境安全的影响

IF 8.2 Q1 AGRICULTURE, MULTIDISCIPLINARY
Leo Ramos , Edmundo Casas , Eduardo Bendek , Cristian Romero , Francklin Rivas-Echeverría
{"title":"用于烟雾和野火探测的 YOLOv8 超参数优化:对农业和环境安全的影响","authors":"Leo Ramos ,&nbsp;Edmundo Casas ,&nbsp;Eduardo Bendek ,&nbsp;Cristian Romero ,&nbsp;Francklin Rivas-Echeverría","doi":"10.1016/j.aiia.2024.05.003","DOIUrl":null,"url":null,"abstract":"<div><p>In this study, we extensively evaluated the viability of the state-of-the-art YOLOv8 architecture for object detection tasks, specifically tailored for smoke and wildfire identification with a focus on agricultural and environmental safety. All available versions of YOLOv8 were initially fine-tuned on a domain-specific dataset that included a variety of scenarios, crucial for comprehensive agricultural monitoring. The ‘large’ version (YOLOv8l) was selected for further hyperparameter tuning based on its performance metrics. This model underwent a detailed hyperparameter optimization using the One Factor At a Time (OFAT) methodology, concentrating on key parameters such as learning rate, batch size, weight decay, epochs, and optimizer. Insights from the OFAT study were used to define search spaces for a subsequent Random Search (RS). The final model derived from RS demonstrated significant improvements over the initial fine-tuned model, increasing overall precision by 1.39 %, recall by 1.48 %, F1-score by 1.44 %, [email protected] by 0.70 %, and [email protected]:0.95 by 5.09 %. We validated the enhanced model's efficacy on a diverse set of real-world images, reflecting various agricultural settings, to confirm its robustness in detecting smoke and fire. These results underscore the model's reliability and effectiveness in scenarios critical to agricultural safety and environmental monitoring. This work, representing a significant advancement in the field of fire and smoke detection through machine learning, lays a strong foundation for future research and solutions aimed at safeguarding agricultural areas and natural environments.</p></div>","PeriodicalId":52814,"journal":{"name":"Artificial Intelligence in Agriculture","volume":null,"pages":null},"PeriodicalIF":8.2000,"publicationDate":"2024-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S2589721724000187/pdfft?md5=c551b82b80431a9f2f37f79894497fcb&pid=1-s2.0-S2589721724000187-main.pdf","citationCount":"0","resultStr":"{\"title\":\"Hyperparameter optimization of YOLOv8 for smoke and wildfire detection: Implications for agricultural and environmental safety\",\"authors\":\"Leo Ramos ,&nbsp;Edmundo Casas ,&nbsp;Eduardo Bendek ,&nbsp;Cristian Romero ,&nbsp;Francklin Rivas-Echeverría\",\"doi\":\"10.1016/j.aiia.2024.05.003\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><p>In this study, we extensively evaluated the viability of the state-of-the-art YOLOv8 architecture for object detection tasks, specifically tailored for smoke and wildfire identification with a focus on agricultural and environmental safety. All available versions of YOLOv8 were initially fine-tuned on a domain-specific dataset that included a variety of scenarios, crucial for comprehensive agricultural monitoring. The ‘large’ version (YOLOv8l) was selected for further hyperparameter tuning based on its performance metrics. This model underwent a detailed hyperparameter optimization using the One Factor At a Time (OFAT) methodology, concentrating on key parameters such as learning rate, batch size, weight decay, epochs, and optimizer. Insights from the OFAT study were used to define search spaces for a subsequent Random Search (RS). The final model derived from RS demonstrated significant improvements over the initial fine-tuned model, increasing overall precision by 1.39 %, recall by 1.48 %, F1-score by 1.44 %, [email protected] by 0.70 %, and [email protected]:0.95 by 5.09 %. We validated the enhanced model's efficacy on a diverse set of real-world images, reflecting various agricultural settings, to confirm its robustness in detecting smoke and fire. These results underscore the model's reliability and effectiveness in scenarios critical to agricultural safety and environmental monitoring. This work, representing a significant advancement in the field of fire and smoke detection through machine learning, lays a strong foundation for future research and solutions aimed at safeguarding agricultural areas and natural environments.</p></div>\",\"PeriodicalId\":52814,\"journal\":{\"name\":\"Artificial Intelligence in Agriculture\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":8.2000,\"publicationDate\":\"2024-06-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://www.sciencedirect.com/science/article/pii/S2589721724000187/pdfft?md5=c551b82b80431a9f2f37f79894497fcb&pid=1-s2.0-S2589721724000187-main.pdf\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Artificial Intelligence in Agriculture\",\"FirstCategoryId\":\"1087\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S2589721724000187\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"AGRICULTURE, MULTIDISCIPLINARY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Artificial Intelligence in Agriculture","FirstCategoryId":"1087","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2589721724000187","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"AGRICULTURE, MULTIDISCIPLINARY","Score":null,"Total":0}
引用次数: 0

摘要

在本研究中,我们广泛评估了最先进的 YOLOv8 架构在物体检测任务中的可行性,该架构专门针对烟雾和野火识别而定制,重点关注农业和环境安全。YOLOv8 的所有可用版本最初都是在特定领域的数据集上进行微调的,该数据集包括对全面农业监测至关重要的各种场景。根据其性能指标,"大型 "版本(YOLOv8l)被选中进行进一步的超参数调整。该模型使用 "一次一个因素"(OFAT)方法进行了详细的超参数优化,主要集中在学习率、批量大小、权重衰减、历时和优化器等关键参数上。从 OFAT 研究中获得的启示被用于定义后续随机搜索 (RS) 的搜索空间。与最初的微调模型相比,RS 得出的最终模型有了显著改进,总体精确度提高了 1.39%,召回率提高了 1.48%,F1 分数提高了 1.44%,[email protected] 提高了 0.70%,[email protected]:0.95 提高了 5.09%。我们在一组反映不同农业环境的真实图像上验证了增强型模型的功效,以确认其在检测烟雾和火灾方面的鲁棒性。这些结果凸显了该模型在对农业安全和环境监测至关重要的场景中的可靠性和有效性。这项工作代表了机器学习在火灾和烟雾探测领域取得的重大进展,为今后旨在保护农业地区和自然环境的研究和解决方案奠定了坚实的基础。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Hyperparameter optimization of YOLOv8 for smoke and wildfire detection: Implications for agricultural and environmental safety

In this study, we extensively evaluated the viability of the state-of-the-art YOLOv8 architecture for object detection tasks, specifically tailored for smoke and wildfire identification with a focus on agricultural and environmental safety. All available versions of YOLOv8 were initially fine-tuned on a domain-specific dataset that included a variety of scenarios, crucial for comprehensive agricultural monitoring. The ‘large’ version (YOLOv8l) was selected for further hyperparameter tuning based on its performance metrics. This model underwent a detailed hyperparameter optimization using the One Factor At a Time (OFAT) methodology, concentrating on key parameters such as learning rate, batch size, weight decay, epochs, and optimizer. Insights from the OFAT study were used to define search spaces for a subsequent Random Search (RS). The final model derived from RS demonstrated significant improvements over the initial fine-tuned model, increasing overall precision by 1.39 %, recall by 1.48 %, F1-score by 1.44 %, [email protected] by 0.70 %, and [email protected]:0.95 by 5.09 %. We validated the enhanced model's efficacy on a diverse set of real-world images, reflecting various agricultural settings, to confirm its robustness in detecting smoke and fire. These results underscore the model's reliability and effectiveness in scenarios critical to agricultural safety and environmental monitoring. This work, representing a significant advancement in the field of fire and smoke detection through machine learning, lays a strong foundation for future research and solutions aimed at safeguarding agricultural areas and natural environments.

求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
Artificial Intelligence in Agriculture
Artificial Intelligence in Agriculture Engineering-Engineering (miscellaneous)
CiteScore
21.60
自引率
0.00%
发文量
18
审稿时长
12 weeks
×
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
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
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 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信