SafetyStrapAI/README.md

Index

1. Introduction
2. Features
3. Installation
4. Usage
5. Configuration
6. Example
7. Dependencies
8. License
9. Contact

Introduction

A Python-based tool designed to identify whether drivers in a set of images are wearing seat belts. Given a folder containing images, the application processes each image and generates a report indicating which drivers are wearing seat belts and which are not.

Features

• Batch Processing: Processes multiple images in a single run.
• Accurate Detection: Utilizes advanced machine learning models to ensure high accuracy.
• Detailed Reports: Generates comprehensive reports of the results.
• User-Friendly: Simple command-line interface for easy interaction.

Installation

1. Clone the Repository:

   git clone https://github.com/vlalef/SafetyStrapAI.git
   cd SafetyStrapAI

2. Create and Activate a Virtual Environment:

   python -m venv venv
   source venv/bin/activate
   

3. Install Dependencies: Install the dependencies:

   pip install -r requirements.txt
   

Usage

1. Prepare your dataset: Ensure that the images are organized into the appropriate directories under dataset/train and dataset/validation as it follows.

   dataset/ 
   ├── train/
   │   ├── with_seatbelt/
   │   │   ├── 1.jpg
   │   │   ├── 2.jpg
   │   │   ├── 3.jpg
   │   │   ├── 4.jpg
   │   │   ├── 5.jpg
   │   │   ├── 6.jpg
   │   │   ├── 7.jpg
   │   │   ├── 8.jpg
   │   │   ├── 9.jpg
   │   │   └── 10.jpg
   │   └── without_seatbelt/
   │       ├── 1.jpg
   │       ├── 2.jpg
   │       ├── 3.jpg
   │       ├── 4.jpg
   │       ├── 5.jpg
   │       ├── 6.jpg
   │       ├── 7.jpg
   │       ├── 8.jpg
   │       ├── 9.jpg
   │       └── 10.jpg
   └── validation/
       ├── with_seatbelt/
       │   ├── 1.jpg
       │   ├── 2.jpg
       │   ├── 3.jpg
       │   ├── 4.jpg
       │   └── 5.jpg
       └── without_seatbelt/
           ├── 1.jpg
           ├── 2.jpg
           ├── 3.jpg
           ├── 4.jpg
           └── 5.jpg
   

2. Train the model:

   python train_model.py
   

   This command will train the model using the images in the train directory and validate it using images from the validation directory.

3. Run the application:

   python safety_strap.py --input_folder ./dataset/validation --output_folder ./results --model ./models/seatbelt_detector.keras
   

4. Review the Results:

   The application will generate a report in the specified output folder detailing which images contain drivers using seat belts and which do not.

   Example Report:

   image1.jpg: Using seat belt
   image2.jpg: Not using seat belt
   image3.jpg: Using seat belt
   

Configuration

The application can be configured via command-line arguments:

• --input_folder: Path to the directory containing the images.
• --output_folder: Path to the directory where results will be saved.
• --model: (Optional) Path to a custom model file if you wish to use a different model.

Example:

python safety_strap.py --input_folder ./dataset/validation --output_folder ./results --model ./models/seatbelt_detector.keras

Dependencies

• Python 3.9+
• OpenCV
• TensorFlow
• NumPy
• Keras
• Pillow
• SciPy

License

This project is licensed under the GPL-3.0 license. Se the LICENSE file for more details

Contact

• Author: Alef L. Vaz
• Email: [email protected]
• GitHub: vlalef