ABSTRACT:
Mental health and stress-related issues have become a significant concern in today’s fast-paced lifestyle, affecting individuals’ productivity, emotional well-being, and overall health. Early identification of emotional states and stress levels can help in timely intervention and support. This project presents an AI-based Emotion and Stress Detection System with Real-Time Feedback and Alert Mechanism, designed to monitor a person’s emotional and physiological condition in a non-invasive and user-friendly manner.
The proposed system utilizes a camera-based facial emotion recognition model to identify human emotions such as happiness, sadness, anger, fear, surprise, and neutral state using deep learning techniques. A Convolutional Neural Network (CNN) is employed to analyze facial expressions captured in real time and classify emotional states accurately. In parallel, physiological stress indicators are monitored using a SpO₂ and heart rate sensor (MAX30102) along with a temperature sensor (DHT11). These sensors provide vital parameters such as heart rate, oxygen saturation, and body temperature, which are analyzed to estimate the user’s stress level.
The system integrates both emotion recognition and sensor-based stress analysis to achieve a multimodal stress detection approach, improving reliability compared to single-parameter systems. The detected emotion, stress level, and physiological readings are displayed in real time on an LCD module, enabling immediate visual feedback to the user. Additionally, the system incorporates emotion-based motivational and supportive messages, which are displayed on the LCD to provide psychological encouragement and emotional reassurance based on the detected emotional state.
To enhance safety and remote monitoring, an SMS alert system using Twilio is implemented. When high stress levels are detected, automatic alert messages containing the detected emotion, stress level, and support information are sent to predefined contacts such as caretakers or family members. This feature enables timely awareness and intervention in critical situations.
Overall, the proposed system offers a low-cost, real-time, and intelligent emotional well-being monitoring solution suitable for applications in healthcare monitoring, workplace stress management, student mental health assessment, and personal wellness systems. By combining artificial intelligence, sensor data, and supportive feedback, the system aims not only to detect stress and emotions but also to promote emotional awareness and mental well-being.
INTRODUCTION:
In recent years, mental health and emotional well-being have emerged as critical aspects of overall human health, especially due to increasing academic pressure, workplace stress, social challenges, and lifestyle changes. Stress and emotional imbalance, if left unrecognized, can lead to serious health issues such as anxiety disorders, depression, reduced productivity, and chronic illnesses. Therefore, continuous monitoring of emotional states and stress levels has become an important area of research in healthcare and human–computer interaction.
Traditional methods of stress and emotion assessment mainly rely on self-reporting, questionnaires, or clinical evaluations, which are often subjective, time-consuming, and not suitable for real-time monitoring. With the rapid advancement of Artificial Intelligence (AI), computer vision, and sensor technologies, it has become possible to automatically analyze human emotions and physiological signals in a non-invasive and efficient manner. Facial expressions are one of the most natural and informative indicators of a person’s emotional state, while physiological parameters such as heart rate, oxygen saturation, and body temperature provide valuable insight into stress conditions.
This project proposes an AI-based Emotion and Stress Detection System that combines facial emotion recognition with sensor-based physiological analysis to provide a comprehensive understanding of a person’s mental state. The system uses a camera to capture facial images and applies deep learning techniques to classify emotions such as happiness, sadness, anger, fear, surprise, and neutral state. Simultaneously, physiological data including heart rate, SpO₂ level, and body temperature are collected using appropriate sensors to estimate the stress level of the individual.
To enhance usability and real-world applicability, the system provides real-time feedback through an LCD display, showing detected emotions, stress levels, and physiological readings. In addition to detection, the system delivers emotion-based motivational and supportive messages, helping users feel emotionally reassured and encouraged. Furthermore, an alert mechanism using SMS communication is integrated to notify caregivers or concerned individuals when high stress levels are detected, enabling timely intervention.
The proposed system is designed to be low-cost, non-invasive, and user-friendly, making it suitable for applications such as student stress monitoring, workplace mental health assessment, personal wellness tracking, and basic healthcare support systems. By combining AI-driven emotion analysis with physiological sensing and emotional support, this project aims not only to detect stress and emotions but also to promote awareness, early intervention, and overall emotional well-being.
OBJECTIVE:
The primary objective of this project is to design and develop an AI-based system capable of detecting human emotions and stress levels in real time, using facial expression analysis and physiological sensor data. The specific objectives of the project are as follows:
1. To develop a facial emotion recognition system using a camera and deep learning techniques to identify emotions such as happiness, sadness, anger, fear, surprise, and neutral state.
2. To monitor physiological parameters such as heart rate, oxygen saturation (SpO₂), and body temperature using non-invasive sensors for stress level estimation.
3. To analyze and classify stress levels (normal, mild, or high) based on sensor data using suitable machine learning algorithms.
4. To integrate emotion detection and stress analysis into a multimodal system to improve the accuracy and reliability of mental state assessment.
5. To display real-time emotion, stress level, and physiological readings on an LCD module for immediate user feedback.
6. To provide emotion-based motivational and supportive messages corresponding to the detected emotional state in order to enhance emotional reassurance and user well-being.
7. To implement an alert mechanism using SMS communication to notify caregivers or concerned individuals when high stress levels are detected.
8. To design a low-cost, user-friendly, and non-invasive system suitable for continuous monitoring in real-world environments.
9. To promote early stress detection and emotional awareness, enabling timely intervention and preventive mental healthcare support.