Published May 5, 2026 | Version v1
Journal article Open

Human Activity Recognition

Description

This project presents an intelligent and automated safety monitoring system that integrates Human Action Recognition and Facial Expression Recognition to detect critical emergency situations such as falls and drowsiness in real time. The system continuously analyzes video or motion-based input data to understand both the physical activities and emotional states of individuals in monitored environments, enabling a comprehensive and reliable assessment of human behavior. The Human Action Recognition module identifies common daily activities including walking, running, sitting, standing, lying down, and climbing stairs, with a particular emphasis on accurate fall detection by distinguishing normal movements from hazardous events. Simultaneously, the Facial Expression Recognition module monitors facial features to recognize emotional states such as happiness, sadness, anger, fear, surprise, disgust, and neutral expressions, while also detecting drowsiness through indicators such as abnormal posture, reduced facial responsiveness, prolonged eye closure, or sudden head movements. Continuous camera-based monitoring allows the system to detect abnormal behavior patterns and emergency conditions at an early stage. Upon detecting a fall, drowsiness, or other critical event, the system automatically generates and sends an emergency email alert to predefined contacts, containing essential information such as the type of incident, detected activity or facial expression, exact date and time of occurrence, and the user’s current location in the form of GPS coordinates or a physical address. By integrating multiple AI-based monitoring models into a unified and scalable framework, the proposed system significantly reduces dependence on manual supervision, minimizes human error, and improves emergency response time. The system is well suited for deployment in elderly care and patient monitoring, healthcare and rehabilitation systems, and smart surveillance environments, where continuous, intelligent, and proactive safety monitoring is essential for enhancing overall safety and quality of life.

Files

17-M Manoj Kumar.pdf

Files (1.0 MB)

Name Size Download all
md5:f97ce3f018201ea8ecdee1b3f65f57c9
1.0 MB Preview Download