2024-25

Permanent URI for this collection

http://192.168.75.5:4000/handle/123456789/17278

Browse

Browsing 2024-25 by Author "Anush L 1NH21EE016, Kruthika B J1NH21EE047, D Kaveri Isha1NH21IS045, P Pawar1NH21IS064"

Now showing 1 - 1 of 1
  • Thumbnail Image
    Item
    Health monitoring system with CNN Based facial emotion detection
    (NHCE, 2024) Anush L 1NH21EE016, Kruthika B J1NH21EE047, D Kaveri Isha1NH21IS045, P Pawar1NH21IS064
    The rapid advancement of wearable and smart sensing technology has enabled innovative systems for real-time health monitoring. This project presents a health monitoring system integrated with a CNN-based facial emotion detection module, designed to assess both physiological and psychological states. The system combines multiple components to provide comprehensive insights into an individual's well-being. The health monitoring module uses an Arduino Uno microcontroller connected to sensors for measuring heart rate and body temperature. A pulse sensor detects heart rate in beats per minute (BPM) by analyzing changes in blood flow, while an LM35 temperature sensor measures body temperature in degrees Celsius. These physiological parameters are transmitted to a connected computer via a USB interface and processed using Python to analyze and visualize the data. The emotion detection module utilizes a Convolutional Neural Network. (CNN) trained to classify facial expressions into categories such as happiness, sadness, anger, surprise, and neutrality. A camera captures real-time video, and the system detects and processes facial regions to identify emotions. The emotion detection module is critical for understanding an individual's mental state, as emotional shifts often correlate with stress, anxiety, or other health conditions. To enhance functionality, the system integrates emotion detection and physiological data. For example, sudden emotional changes (like fear or anger) can be cross-referenced with spikes in heart rate and variations in temperature to assess stress levels or potential health risks. This integration enables a more holistic view of health, combining physical and emotional indicators. The proposed system has applications in healthcare, fitness, and lie detection, providing continuous monitoring and analysis in real-time. Its usability extends to remote health diagnostics, early detection of stress or anxiety disorders, and assisting healthcare professionals in understanding