Sentiment Analysis on Human with special Concentration on infants’ emotional face using Machine Learning Techniques.
Download as PPTX, PDF1 like101 views
This seminar presentation summarizes a project on sentiment analysis and emotion detection in infants using machine learning techniques. The project aims to develop software to detect emotions like happiness, sadness, anger, disgust, surprise and fear from facial expressions of children and infants using OpenCV and machine learning. The system will include face detection, feature extraction of regions like eyes and mouth, and emotion classification using a neural network trained on a dataset of facial expressions. The application could help provide appropriate care to infants by recognizing their mental and physiological states.
![Background study
• Wathsala Nayomi widanagamaachchi[1] in his research ‘Facial
Emotion Recognition with a Neural Network Approach’ established
the fact that Neural Network provides better accuracy than Naive
Bayes classifier, which is a probabilistic classifier based on Bayes
Theorem.
• Andreea Pascu under guidance of Prof. Ross King[3] in there research
‘Facial Recognition System’ found a strong accuracy of 86% using
Machine Learning.](https://image.slidesharecdn.com/problemdefination-180929105503/85/Sentiment-Analysis-on-Human-with-special-Concentration-on-infants-emotional-face-using-Machine-Learning-Techniques-5-320.jpg)
![References
[1] Wathsala Nayomi Widanagamaachchi. Facial Emotion Recognition with a Neural
Network Approach. 2009.
[2] G.Hemalatha and C.P Samathi, ‘A Study of Techniques for Facial Detection and
Expression Classification’, IJCSES Vol.5, April 2014
[3] Andreea Pascu and Prof. Ross King, ‘Facial Expression Recognition System’, University of
Manchester, April 2015.
[4] M. Pantic, L. J. M. Rothkrantz, and H. Koppelaar. Automation of nonverbal
communication of facial expressions. In in: EUROMEDIA 98, SCS International, pages 86{93,
1998.
[5] T. Pham and M. Worring. Face detection methods: A critical evaluation. ISIS Technical
Report Series, University of Amsterdam, 11, 2000.
[6] T. V. Pham, M. Worring, and A. W. M. Smeulders. Face detection by aggregated bayesian
network classifiers. Pattern Recogn. Lett., 23(4):451{461, 2002.](https://image.slidesharecdn.com/problemdefination-180929105503/85/Sentiment-Analysis-on-Human-with-special-Concentration-on-infants-emotional-face-using-Machine-Learning-Techniques-15-320.jpg)
More Related Content
Similar to Sentiment Analysis on Human with special Concentration on infants’ emotional face using Machine Learning Techniques. (20)
More from Takrim Ul Islam Laskar (6)
Recently uploaded (20)
Sentiment Analysis on Human with special Concentration on infants’ emotional face using Machine Learning Techniques.
- 1. Seminar Presentation on Sentiment Analysis on Human with special Concentration on infants’ emotional face using Machine Learning Techniques. Presented By- Takrim Ul Islam Laskar Roll No. 170302005 M.Tech 3rd Sem Gauhati University Institute of Science And Technology Dept. of Information Technology,Gauhati University
- 2. Contents • Introduction • Problem Statement • Background Study • Motivation • Applications • Action Plan • Tools, Environment and Experimental Plateform • Emotion Detection System • Face Detection Stage • Feature Extraction stage • Emotion Classification Stage • Conclusion • References
- 3. Introduction Emotion is psychological and physiological state which is subjective to different conditions. Broadly classifying happyness, sadness, anger, disgust, surprise and fear are some of them. We are to devolope a software to detect the emotion of child/infant using openCV and Machine Learning. The project falls under the domains Artificial Intellegence and Digital Image Processing.
- 4. Problem Statement Pose, speech, facial expression, behaviour, etc convey emotions of individuals . There’s an old proverb “Face is the index of mind”. All the inner emotions reflect on the face of an individual . So the idea is to read the face and detect the emotional state of the subject as it is easily perceptible and it has higher accuracy. In this project our focus would be to do the same with special concentration on children and infants.
- 5. Background study • Wathsala Nayomi widanagamaachchi[1] in his research ‘Facial Emotion Recognition with a Neural Network Approach’ established the fact that Neural Network provides better accuracy than Naive Bayes classifier, which is a probabilistic classifier based on Bayes Theorem. • Andreea Pascu under guidance of Prof. Ross King[3] in there research ‘Facial Recognition System’ found a strong accuracy of 86% using Machine Learning.
- 6. Motivation • Projects have been done earlier on Emotion Detection with precision. But we have taken up the task of focusing more on infants and children who cannot talk to express their feelings .
- 7. Applications • This can be used for medical purposes to detect infants’ emotions. • This can be used as a facial language recognition system.
- 8. Action Plan Date Milestone End September 2018 Problem Statement End October 2018 Literature review & Preprocessing End December 2018 Feature extraction & Phase 1 Project Report End of march 2019 Data set training Data set testing End of May 2019 Implementation & Phase 2 Project Report Face recognition system can also be collected and proceeded to feature extraction.
- 9. Tools/Environment/Experimental Plateform • Tools – openCV 3.1.0, Python 2.7.12, Artificial Neural Network. • Operating System- Linux based plateform (e.g Ubuntu 16.04 ). • Hardware- RAM 4 GB, Intel Core i5 7th Gen.
- 10. Emotion Detection System Fig 1. Emotion Detection System Stages in Emotion Detection: 1. Face Detection Stage. 2. Feature Extraction Stage. 3. Emotion Classification Stage.
- 11. Face Detection Feature invariant Approach would be implemented . • It would work based on the skin colour. • It is in focus currently as skin colour resides in a small colour range in different colour spaces irrespective of race.
- 12. Feature Extraction Stage 1. Feature Region Extraction: From the face detected, further processing gives eye, eyebrows, mouth, etc. 2. Feature Point Extraction From the regions the corner points are extracted .
- 13. Emotion Classification Stage The feature points are then inputed to the neural network. Neural Network would be used to train a Data Set. This Data Set would be used to classify the emotions inputed. In doing so, a huge amount of facial expressions of different individuals would have to be collected.
- 14. Conclusion The application would be able to detect the mental and physiological state of an individual especially the infant . This can be used to provide appropriate care to the infant.
- 15. References [1] Wathsala Nayomi Widanagamaachchi. Facial Emotion Recognition with a Neural Network Approach. 2009. [2] G.Hemalatha and C.P Samathi, ‘A Study of Techniques for Facial Detection and Expression Classification’, IJCSES Vol.5, April 2014 [3] Andreea Pascu and Prof. Ross King, ‘Facial Expression Recognition System’, University of Manchester, April 2015. [4] M. Pantic, L. J. M. Rothkrantz, and H. Koppelaar. Automation of nonverbal communication of facial expressions. In in: EUROMEDIA 98, SCS International, pages 86{93, 1998. [5] T. Pham and M. Worring. Face detection methods: A critical evaluation. ISIS Technical Report Series, University of Amsterdam, 11, 2000. [6] T. V. Pham, M. Worring, and A. W. M. Smeulders. Face detection by aggregated bayesian network classifiers. Pattern Recogn. Lett., 23(4):451{461, 2002.
- 16. Thank You