PPE Detection Project

This project aims to detect Personal Protective Equipment (PPE) using YOLOv11 or other YOLO models of your choice. It includes data preprocessing, model training, and evaluation to ensure accurate detection of PPE in images.

---

Project Structure

Main Directory
│   .env
│   .gitignore
│   requirements.txt
│   README.md
│
├── dataset
│   ├── dataset_train.yaml
│   ├── images
│   │   ├── test
│   │   ├── train
│   │   └── val
│   └── labels
│       ├── test
│       ├── train
│       └── val
│
├── P1 - Data Preprocessing
│   ├── data_preprocessing.py  # Main script to preprocess images and labels
│   ├── convert_xml_to_txt.py  # Converts annotation files from XML (Pascal VOC) to TXT (YOLO format)
│   ├── display_labels.py      # Displays dataset labels and their distributions
│   └── separate_non_annotated_images.py  # Filters out images without annotations
│
├── P2 - Model Training
│   ├── yolo_v11_model_training.py  # Script for training YOLOv11 model
│   ├── model_loader.py  # Loads and initializes the YOLO model
│   └── augment_data.py  # Applies data augmentation techniques to improve model performance
│
└── P3 - Model Evaluation
    ├── model_evaluation.py  # Evaluates the trained YOLO model using test data
    ├── model_visual_test.py  # Visualizes model predictions on sample images
    └── performance_metrics.py  # Computes key performance metrics (mAP, precision, recall)

---

Setup Instructions

1. Clone the Repository

git clone https://github.com/akshaysatyam2/yolo-for-detection.git
cd yolo-for-detection

2. Create and Activate a Virtual Environment

python -m venv venv

• Windows:

  venv\Scripts\activate

• macOS/Linux:

  source venv/bin/activate

3. Install Required Dependencies

pip install -r requirements.txt

4. Set Up Environment Variables

• Rename .env.example to .env and configure necessary variables.

---

Usage Guide

1. Data Preprocessing

Preprocess the dataset before training.

Run:

python "P1 - Data Preprocessing/data_preprocessing.py"

2. Model Training

Train the YOLOv11 model on the dataset.

Run:

python "P2 - Model Training/yolo_v11_model_training.py"

3. Model Evaluation

Evaluate the trained model’s performance.

Run:

python "P3 - Model Evaluation/model_evaluation.py"

---

Dataset Structure

The dataset follows this format:

dataset/
├── images/
│   ├── test/
│   ├── train/
│   └── val/
└── labels/
    ├── test/
    ├── train/
    └── val/

• Images: Store images in images/train, images/val, and images/test.
• Labels: Store corresponding labels in labels/train, labels/val, and labels/test.

---

Configuration Files

Yaml Configuration (dataset_train.yaml)

train: dataset/images/train
val: dataset/images/val
test: dataset/images/test
nc: 2
names: ["Helmet", "No Helmet"]
epochs: 50
batch_size: 16
image_size: 640

---

Required Dependencies

This project requires:

• ultralytics
• opencv-python
• numpy
• matplotlib
• pandas
• python-dotenv

Install with:

pip install -r requirements.txt

---

Deactivating the Virtual Environment

When done, deactivate the virtual environment:

deactivate

---

Contact Information

For questions or contributions:

• Name: Akshay Kumar
• Email: akshaysatyam2003@gmail.com
• GitHub: @akshaysatyam2

---