FACE DETECTION AND EMOTION

RECOGNITION
SUBHADIP CHAKRABORTY, AMIT MANDAL, RAGHWENDRA KUMAR,
BANASHREE CHATTERJEE,
GROUP NO: 34, DEPARTMENT: INFORMATION TECHNOLOGY

ABSTRACT METHODOLOGY RESULT

This project aims to develop a Our approach begins with collecting
comprehensive system for face diverse image datasets depicting
detection and gesture individuals expressing various Our facial detection and emotion recognition system
recognition, enhancing user emotions. Using OpenCV's Haar achieved an impressive accuracy rate of 90% in
interaction with digital interfaces. cascade classifier, we conduct face detecting faces and classifying emotions in controlled
Leveraging advancements in deep detection to localize facial regions testing environments. However, we acknowledge the
learning, our system employs within the images. DeepFace, a pre- ongoing efforts to further enhance its accuracy and
open cv, matplotlib, scikitlearn, trained model, is then employed for robustness. Through continuous refinement and
and numpy, CNN for real-time emotion recognition, analyzing the optimization of our algorithms, we aim to improve the
gesture interpretation in various facial expressions within the system's performance and extend its capabilities for
domains, from healthcare to detected regions to classify real-world applications. Despite the significant progress
entertainment. emotions. Integration of face
METHODOLOGY R
achieved, we remain committed to advancing the
detection and emotion recognition E Discuss and analysis of the results of your project.
S
accuracy of our system to ensure its reliability across
Insert picture figure of the result. It may be graph,
outputs leads to a unified system. diverse scenarios and contexts.
table, photographs of prototype etc. Discuss and
Evaluation metrics such as accuracy U analysis of the results of your project. Insert pic-
INTRODUCTION and precision are employed,
L ture figure of the result. It may be graph, table,
T | Accuracy Type | Accuracy Level |
photographs of prototype etc. Discuss and anal-
followed by optimization techniques ysis of the results of your project. Insert picture
|---------------|----------------|
figure of the result. It may be graph, table, pho-
to enhance system performance. tographs of prototype etc.
In today's digital era, Figure 1: Figure caption| Train | 92% |
INTRODUCTION Real-world validation tests the | Test | 85% |
understanding human emotions system's ability to detect faces and
through facial cues has gained classify emotions in diverse
immense importance across conditions, ensuring its
various fields. This project aims to effectiveness. Through this
develop a robust system for facial streamlined methodology, we
detection and emotion develop a robust system for facial
recognition by leveraging detection and emotion recognition.
advanced technologies such as CONCLUSION
OpenCV and DeepFace.Through
the fusion of Python, NumPy, and Data collection
Matplotlib, CNN we endeavor to and Face detection
create a sophisticated solution preprocessing
capable of accurately detecting
faces and deciphering the
emotions they express. By
harnessing the power of these
technologies, we seek to bridge Emotion
CONCLUSION
Intregation
the gap between artificial recognition
intelligence
REFERENCES
and human emotion
understanding. • - Gathered diverse datasets for training and testing
This project represents a facial detection and emotion recognition algorithms.
significant step forward in human- • - Preprocessed images to enhance quality and
computer interaction and mental consistency.
health assessment, offering Evaluation Optimization • - Employed OpenCV and DeepFace technologies for
insights into the emotional states accurate face detection and emotion classification.
of individuals through facial • - Evaluated system performance, achieving 90%
analysis. Our ultimate goal is to accuracy in controlled environments.
contribute to the development of • - Continued efforts directed towards improving
more empathetic and intelligent accuracy and applicability in real-world scenarios.
systems in the digital realm.
REFERENCES
• Swain, M., Routray, A., Kabisatpathy, P.: Databases, • Facial Emotion Recognition: A Survey of Deep Learning
features and classifiers for speech emotion recognition: a Techniques" by S. Kumari and M. K. Rana (2020).
review. Int. J. Speech Technol. 21(1), 93–120 (2018)
• A Review on Facial Emotion Recognition Techniques Using Deep
• Ko, B.C.: A brief review of facial emotion recognition Learning" by M. S. Hossain, M. I. Hussain, and A. Al-Qurishi
based on visual information. Sensors 18(2), 401 (2018) (2021).