Scholarlink

<p>Natural disasters can cause destruction and harm to people and property. This paper proposes a human detection system using drones and robots implemented through the Internet of Things (IoT) to aid rescue efforts. This model uses the Haar Cascade algorithm to detect humans and extract essential information from camera-captured images, such as facial and eye features. The model takes the current frame captured and compares the frame with the database to find similarities in gradient to identify human body parts. The processed images and environmental data from sensors are sent to the cloud for storage. This system can help in faster and more efficient rescue missions by enabling rescuers to focus only on rescue operations and not waste time searching for victims. In addition, a mobile application is created to visualize the victim’s information, which can aid the rescue in time and effort. With this approach, 91% prediction accuracy was acheived in detecting humans while offering advantages in terms of power efficiency and the compact robot size.</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

IoT-based human detection for fast earthquake response and rescue

  • 期刊论文

  • 发表日期：

    2024年12月31日

查看全文
  • S. Vaideeshwaran,
  • Aditya Jaganath,
  • Anish Subramanian,
  • G. Seeja,
  • V. Berlin Hency

摘要

Natural disasters can cause destruction and harm to people and property. This paper proposes a human detection system using drones and robots implemented through the Internet of Things (IoT) to aid rescue efforts. This model uses the Haar Cascade algorithm to detect humans and extract essential information from camera-captured images, such as facial and eye features. The model takes the current frame captured and compares the frame with the database to find similarities in gradient to identify human body parts. The processed images and environmental data from sensors are sent to the cloud for storage. This system can help in faster and more efficient rescue missions by enabling rescuers to focus only on rescue operations and not waste time searching for victims. In addition, a mobile application is created to visualize the victim’s information, which can aid the rescue in time and effort. With this approach, 91% prediction accuracy was acheived in detecting humans while offering advantages in terms of power efficiency and the compact robot size.