Scholarlink

<p>Difficulties in effectively identifying and categorizing fires, particularly minor ones, within extensive forested areas globally. Due to detecting difficulties and hardware deployment challenges in remote sites, traditional approaches have difficulty implementing early warning systems in such circumstances. The study suggests machine learning algorithm based on Computer Vision model<b> (</b>YOLOv5), a cutting-edge object identification system, in conjunction with attention mechanisms as a solution to this problem. With this all-encompassing strategy, we hope to improve the recognition of small fire targets that are essential for early warning systems in forest areas. Computer Vision model<b> (</b>YOLOv5) achieved remarkable accuracy measures, such as a precision of 98.6%, recall of 90.2%, and an exceptional F1-score of 96%, indicating that the integration produced promising outcomes. The findings indicate noteworthy progress in the precision of detection, which is crucial for efficient handling of forest fires and prompt action to reduce any harm to property and casualties and loss of life. This kind of study will leads to improve fire safety and it will indirectly helps in effective disaster preparedness.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Fire and smoke detection using adoption of machine learning algorithm for improving fire safety and disaster preparedness

  • 期刊论文

  • 发表日期：

    2025年5月26日

查看全文
  • S. Selvakumara Samy,
  • Y. Sai Swarup,
  • T. Sujith Kumar,
  • C. Lakshmi Mani Shankar,
  • S. Krishna Pradeep Reddy,
  • J. S. Sudarsan,
  • S. Nithiyanantham

摘要

Difficulties in effectively identifying and categorizing fires, particularly minor ones, within extensive forested areas globally. Due to detecting difficulties and hardware deployment challenges in remote sites, traditional approaches have difficulty implementing early warning systems in such circumstances. The study suggests machine learning algorithm based on Computer Vision model (YOLOv5), a cutting-edge object identification system, in conjunction with attention mechanisms as a solution to this problem. With this all-encompassing strategy, we hope to improve the recognition of small fire targets that are essential for early warning systems in forest areas. Computer Vision model (YOLOv5) achieved remarkable accuracy measures, such as a precision of 98.6%, recall of 90.2%, and an exceptional F1-score of 96%, indicating that the integration produced promising outcomes. The findings indicate noteworthy progress in the precision of detection, which is crucial for efficient handling of forest fires and prompt action to reduce any harm to property and casualties and loss of life. This kind of study will leads to improve fire safety and it will indirectly helps in effective disaster preparedness.