Skip to content

Latest commit

 

History

History
142 lines (105 loc) · 3.81 KB

README.md

File metadata and controls

142 lines (105 loc) · 3.81 KB
Raw

BMI-Detection-with-Facial-Features

This project uses a deep learning model to predict a person's BMI (Body Mass Index) and gender based on two input images: a front-view and a side-view image. The model leverages a pre-trained VGG16 network for feature extraction and has been fine-tuned for this specific task.

Table of Contents

Overview

This project predicts:

  1. BMI (Body Mass Index): A numerical value indicating whether a person is underweight, normal, or overweight.
  2. Gender: Categorized as male or female based on the input images.

The model processes two images: a front view and a side view. Predictions are based on features extracted using a convolutional neural network.

Features

  • Two-Image Input: Utilizes front and side view images for accurate prediction.
  • BMI Categorization: Categorizes BMI as:
    • Underweight
    • Normal
    • Overweight
  • User-Friendly Interface: Supports selecting images using a graphical file dialog.
  • Expandable: The model can be further trained or integrated into larger systems.

Installation

Prerequisites

  • Python 3.7 or later
  • Required Python libraries:
    • TensorFlow
    • NumPy
    • Pillow
    • Tkinter (for GUI)

Steps

  1. Clone this repository: 
    git clone https://github.com/your-username/BMI-detection-with-facial-features.git
cd BMI-detection-with-facial-features
  2. Install dependencies: 
    pip install -r requirements.txt
  3. Download the trained model and save it to the appropriate path: 
    C:\Users\azadk\OneDrive\Desktop\projects\bmi_detection\log\model_epoch_07_val_loss_29.54.keras

Usage

  1. Run the script:

    python bmi_gender_prediction.py

  2. A dialog will prompt you to select two images:

    • Front View Image
    • Side View Image

  3. The script will display:

    • Predicted BMI
    • BMI Category (Underweight, Normal, Overweight)
    • Predicted Gender (Male or Female)

  4. Example Output:

    A02244 A02244

    Predicted BMI: 23.45
Predicted Gender: Male
BMI Category: Over Weight

Output Results

After running the script, the output is displayed in the terminal. Here’s a breakdown of what each output represents:

  1. Predicted BMI:

    • A numerical value, such as 23.45, representing the predicted Body Mass Index.

  2. BMI Category:

    • Describes the category based on the BMI:
      • Underweight: BMI < 18.5
      • Normal: 18.5 ≤ BMI < 25
      • Overweight: BMI ≥ 25

    image

Model Details

  • Architecture: Based on VGG16 for feature extraction.
  • Input Size: 224x224 pixels for both front and side view images.
  • Outputs:
    • BMI (continuous value)
    • Gender (binary classification: Male/Female)

Limitations

  • Image Quality: Low-resolution or poorly-lit images may affect predictions.
  • Bias: The model's accuracy may vary based on the dataset it was trained on.
  • Generalization: Predictions may not generalize well to certain populations or body types not represented in the training data.

Contributing

Contributions are welcome! To contribute:

  1. Fork the repository.
  2. Create a new branch: 
    git checkout -b feature-branch
  3. Commit your changes: 
    git commit -m "Description of changes"
  4. Push to your branch and create a pull request.