Scholarlink

<p>Commonly found in environmental monitoring systems, PIR (Passive Infrared Sensors) sensors generate a dependable trigger signal when they detect the presence of people or animals. Nevertheless, several factors other than the presence of people or animals affect the PIR sensor’s output. These include the body’s distance from the sensor, its direction of motion, and its kind (human or animal, for example). The purpose of this study is to provide a low-cost PIR sensor classification algorithm that distinguishes between humans and animals and is suitable for intrusion detection systems in outdoor contexts. This is achieved by taking a PIR sensor’s analog signal and extracting time-domain information like signal duration and peak-to-peak amplitude. The developed system can not only classify humans and animals, but it can also estimate the distance of the intruder from the sensor and its direction of movement. The accuracy for the classification of humans and animals and the distance estimation is 95%.</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

PIR Sensor Based Intruder Detection with Direction and Distance Estimation Supported by SVM

  • 期刊论文

  • 发表日期：

    2025年3月11日

查看全文
  • Durlav Sonowal,
  • Anupam Sharma,
  • Bharati Chetia,
  • Nimisha Dutta,
  • Rajdeep Mazumdar,
  • Ananya Bonjyotsna,
  • Kaushik Dehingia

摘要

Commonly found in environmental monitoring systems, PIR (Passive Infrared Sensors) sensors generate a dependable trigger signal when they detect the presence of people or animals. Nevertheless, several factors other than the presence of people or animals affect the PIR sensor’s output. These include the body’s distance from the sensor, its direction of motion, and its kind (human or animal, for example). The purpose of this study is to provide a low-cost PIR sensor classification algorithm that distinguishes between humans and animals and is suitable for intrusion detection systems in outdoor contexts. This is achieved by taking a PIR sensor’s analog signal and extracting time-domain information like signal duration and peak-to-peak amplitude. The developed system can not only classify humans and animals, but it can also estimate the distance of the intruder from the sensor and its direction of movement. The accuracy for the classification of humans and animals and the distance estimation is 95%.