Scholarlink

Over the past two decades, significant technological advancements have taken place, giving rise to novel applications of intelligent digital sensors. The proliferation of Internet of Things (IoT) services in various settings such as smart homes, buildings, cities, factories, and other intelligent environments has brought substantial benefits to individuals, industries, and public institutions. A typical smart home has undergone enhancements and is equipped with an array of sensors and actuators to provide services to its residents. One of the key aspects in many smart home applications is the identification of residents’ daily routines. Automating the Human Activity Recognition system based on human behavior patterns is a complex task due to the intricate nature of human life within a home environment, where multiple inhabitants and residents coexist. To address this complexity, various deep learning algorithms, known for their effectiveness in diverse domains, have been explored to enhance the accuracy of unveiling the human activity pattern. This research aims to investigate and develop a Human Activity Recognition system for smart cities through the implementation of a hybrid deep learning approach known as CNN-LSTM.

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

Unveiling Human Activity Patterns in Smart Cities Through a CNN-LSTM Approach

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Imane Ameur,
  • Mohamed El Amine Ameur,
  • Mazene Ameur,
  • Houssem Eddine Dagha

摘要

Over the past two decades, significant technological advancements have taken place, giving rise to novel applications of intelligent digital sensors. The proliferation of Internet of Things (IoT) services in various settings such as smart homes, buildings, cities, factories, and other intelligent environments has brought substantial benefits to individuals, industries, and public institutions. A typical smart home has undergone enhancements and is equipped with an array of sensors and actuators to provide services to its residents. One of the key aspects in many smart home applications is the identification of residents’ daily routines. Automating the Human Activity Recognition system based on human behavior patterns is a complex task due to the intricate nature of human life within a home environment, where multiple inhabitants and residents coexist. To address this complexity, various deep learning algorithms, known for their effectiveness in diverse domains, have been explored to enhance the accuracy of unveiling the human activity pattern. This research aims to investigate and develop a Human Activity Recognition system for smart cities through the implementation of a hybrid deep learning approach known as CNN-LSTM.