Ethiopian Athletes Image Classifier

In this project, I developed an image classifier for Ethiopian athletes using scikit-learn. I utilized matplotlib for data visualization and used Haar cascade to detect and select images with facial features such as the face, eyes, nose, and mouth. I also used wavelet transformation for feature engineering. This project was built on Google Colab.

Features

• Classifies images of Ethiopian athletes
• Uses Haar cascade for facial feature detection
• Employs wavelet transformation for feature engineering
• Visualizes data using matplotlib

Technologies Used

• Python
• Scikit-learn
• Matplotlib
• OpenCV (for Haar cascade)
• PyWavelets (for wavelet transformation)
• Google Colab

Installation

1. Clone the repository:

   git clone https://github.com/saleamlakw/Ethiopian_athletes_image_classifier.git
   cd Ethiopian_athletes_image_classifier

2. Open the project in Google Colab by uploading the notebook file (ethiopian_athletes_classifier.ipynb).

Usage

1. Upload your dataset to Google Colab or use the provided dataset.
2. Run the cells in the Colab notebook to execute the code step-by-step.
3. The notebook includes data visualization, image preprocessing, feature engineering, and model training.
4. After training, the model can classify new images of Ethiopian athletes.

Data Preprocessing

• Image Selection: Used Haar cascade to detect and select images with prominent facial features such as the face, eyes, nose, and mouth.
• Feature Engineering: Applied wavelet transformation to extract relevant features from the images.

Model Training

• Algorithm: Utilized various algorithms available in scikit-learn.
• Evaluation: Evaluated the model using appropriate metrics and visualized the results using matplotlib.

Visualization

• Matplotlib: Used for plotting graphs and visualizing data distributions and model performance.

Dependencies

The project requires the following Python packages:

scikit-learn
matplotlib
opencv-python
PyWavelets

To install these packages, run:

pip install scikit-learn matplotlib opencv-python PyWavelets