Stress Detection in Facial Expression by Machine Learning
International Journal of Advanced Research in Science Communication and Technology · 29.07.2026 · 657
DOI: 10.48175/ijarsct-37872 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Advanced Research in Science Communication and Technology, online erschienen am 29.07.2026. Verfasst von Harshitha C S und 2 weiteren Autor:innen. Die Arbeit nennt 12 Literaturangaben.
Abstract (Originalsprache)
In today's fast-paced world, stress has become a common problem that affects people of all age groups. Long working hours, academic pressure, and personal challenges can lead to increased stress, which may negatively affect both physical and mental health. Traditional methods of stress assessment mainly depend on questionnaires, interviews, or expert observation. These methods are time-consuming, require human involvement, and may not always provide consistent results. Therefore, there is a need for an automated system that can detect stress quickly and accurately. The proposed system presents a web-based Stress Detection in Facial Expression Using Machine Learning application that analyzes facial expressions to estimate a user's stress level. The system uses image processing techniques to detect the face and extract important facial features. A Convolutional Neural Network (CNN) model is trained to recognize facial emotions such as Happy, Sad, Angry, Fear, Neutral, Surprise, and Disgust. Based on the detected emotion, the system predicts the corresponding stress level and displays the result along with a confidence score. The application follows a simple workflow that includes user authentication, image upload or webcam capture, face detection, image preprocessing, emotion recognition, stress prediction, and report generation. OpenCV is used for face detection and image preprocessing, while TensorFlow and Keras are used to develop the CNN model. The complete application is implemented using Python, Flask, HTML, CSS, JavaScript, and SQLite, providing a user-friendly interface for stress detection. The system was tested using different facial expression images to evaluate its performance. Experimental results showed that the proposed model successfully recognized facial emotions and predicted stress levels with reliable accuracy. The application also stores prediction history and allows users to generate reports for future reference. Although the system is not intended to replace professional medical or psychological evaluation, it can serve as a useful support tool for the early identification of stress in educational institutions, workplaces, healthcare environments, and personal well-being monitoring. Future improvements can further enhance the accuracy and practical use of the proposed system through larger datasets, real-time video analysis, and mobile application support.
Zitieren
APA
S, H. C., Eshwari, B., & R, D. K. (2026). Stress Detection in Facial Expression by Machine Learning. International Journal of Advanced Research in Science Communication and Technology, 657. https://doi.org/10.48175/ijarsct-37872
Vancouver
S HC, Eshwari B, R DK. Stress Detection in Facial Expression by Machine Learning. International Journal of Advanced Research in Science Communication and Technology. 2026:657. doi:10.48175/ijarsct-37872
BibTeX
@article{S2026,
author = {S, Harshitha C and Eshwari, B and R, Dr. Kruthi},
title = {Stress Detection in Facial Expression by Machine Learning},
journal = {International Journal of Advanced Research in Science Communication and Technology},
year = {2026},
pages = {657},
doi = {10.48175/ijarsct-37872}
}
EUVIMED