Emotion-Vision

Artificial intelligence project

Work plan:

1: Data preparation:

• Finding and downloading a suitable dataset containing face images with labels. • Preparing scripts for data processing, such as scaling and rotating images to increase the number of training data.

2: Image processing:

• Applying face detection algorithms such as Haar Cascade Classifier to detect faces in images. • Using data augmentation techniques such as rotation, scaling, brightness adjustment, etc. to generate more training examples.

3: Building CNN model:

• Creating and training a CNN model using machine learning tools such as Keras, TensorFlow, or PyTorch. • Selecting appropriate hyperparameters such as the number of layers, kernel size, and number of filters. • Testing and optimizing the model using techniques such as cross-validation and regularization.

4: Implementation of the user interface:

• Creating a user interface that allows users to upload face images to the CNN model for emotion recognition. • Testing the user interface to ensure it works correctly.

5: Optimization and deployment:

• Optimizing the model to run faster and take up less space.

6: Report:

• Preparing a project report containing a description of the problem, tools and technologies used, implementation description, test results, and conclusions. • Presenting the project results in the form of a presentation to the project group.

Date:

FER2013: dataset containing 35,887 facial images with seven emotion categories (anger, anger, disgust, fear, happiness, sadness, surprise).