Facial Emotion Recognition.pdf

Methodology:

1. Face Detection: Haar Cascade face detection algorithm is used to detect the face from the input image.
2. Gabor Filter: Gabor filter is applied to extract facial features in the spatial domain, reducing computation and size.
3. Feature Extraction: Important facial features are extracted from the facial image after applying the Gabor filter.
4. CNN Model: A Convolutional Neural Network (CNN) model is used to classify facial expressions using the extracted features.

Key Findings:

The proposed method achieves a high recognition rate using the CK+ dataset.
Gabor filter effectively captures the entire frequency spectrum in all directions, improving recognition.
The CNN model successfully classifies facial expressions using the extracted features.

Related Work:

Previous studies have used various methods, including neural networks, decision trees, and support vector machines, to recognize facial emotions.
Gabor filters have been used in other studies to extract features from facial images.
CNN models have been used to classify facial expressions with high accuracy.

Datasets:

CK+ dataset: a widely used dataset for facial emotion recognition.
JAFFE dataset: another dataset used for facial emotion recognition.
Extended Cohn-Kanade dataset: a dataset used for facial emotion recognition with a larger number of images.

Applications:

Facial emotion recognition has applications in game development, market research, and customer feedback analysis.
* It can also be used in fields such as safe driving, medical care, distance education, and marketing assistance.