Download as docx, pdf, or txt
Download as docx, pdf, or txt
You are on page 1of 3

Table Of Contents

Sl No. Topics Page No.

1. Introduction 1
2. Motivation 2

3. Problem Definition 3-4

4. Literature Study 5-11

5. Software Requirement 12

6. Planning 13

7. Design 14-17

8. Algorithm 18

9. Implementation Details 19-30

10. Implementation Of Problem 31-37

11. Result 38-42


12. Conclusion 43

13. Future Scope 44

14. Reference 45
Index Of Images
Sl No. Topics Pages
1. Basic Human Emotion 1
2.1. Monaliza 2
2.2. Deaf & Dumb 2
3. Problem Formulation 3
4.1. Pre-processing 5
4.2. Face Registration 6
4.3. Facial Feature Extraction 6
4.4. Emotion Classification 6
4.5. Neural Network 7
4.6. Gabor Filter 8
5.1. FER2013 Images 19
5.2. FER2013 Sample 19
5.3. Python Alternative To 22
Matlab
5.4. Haar Features 24
5.5. Adaboost 25
5.6. Cascade Workflow 26
5.7. Artificial Neural 28
Network
5.8. Deep Convolution 29
Neural Network
Architecture
5.9. Convolution Neural 30
Network Layers
6.1. Overview Of FER2013 31
Database
6.2. Training & Validation 31
Data Distribution
6.3. FER CNN Architecture 32
6.4. Convolutional & 32
Maxpooling Of Neural
Network
6.5. CNN Forward & 33
Backward Propagation
6.6. Final Model CNN 34
Architecture
6.7. Prediction Of Example 34-35
Faces From Database
6.8. Confusion Matrix 36
6.9. Correct Prediction On 2nd 36
& 3rd Highest Probable
Emotion
6.10. CNN Feature Maps After 37
2nd Layer Of Maxpooling
6.11. CNN Feature Maps After 37
3rd Layer Of
Maxpooling
6.12. Pixel Representation Of 37
Database Images

7.1. Input Sample 1 40


7.2. Greyscale Sample 1 40
7.3 48*48 Greyscale Sample 40
1
7.4 Input Sample 2 41
7.5 Greyscale Sample 2 41
7.6 48*48 Greyscale Sample 41
2
7.7 Input Sample 3 42
7.8 Greyscale Sample 3 42
7.9 48*48 Greyscale Sample 42
3

Index Of Tables
Sl. No. Topics Pages
1. Accuracy of various 10
database
2. Accuracy of various 10-11
approaches are stated
as follows

You might also like