Bicycle Detection with YOLOv10

A computer vision project that uses YOLOv10 (You Only Look Once) to detect bicycles in images and videos. This project includes both pre-trained YOLOv10 models and custom-trained models specifically optimized for bicycle detection.

🚲 Project Overview

This project implements bicycle detection using state-of-the-art YOLOv10 object detection models. It can:

• Detect bicycles in single images
• Process multiple images in batch
• Analyze videos for bicycle detection
• Train custom models on bicycle datasets
• Compare performance between pre-trained and custom models

✨ Features

• Image Detection: Detect bicycles in single images with bounding boxes
• Batch Processing: Process multiple images from a folder
• Video Analysis: Detect bicycles in video files with real-time processing
• Custom Training: Train YOLOv10 models from scratch on bicycle datasets
• Model Comparison: Compare pre-trained YOLOv10 vs custom-trained models
• GPU Support: CUDA-enabled for faster inference and training

🛠️ Requirements

• Python 3.8+
• PyTorch
• Ultralytics
• CUDA-compatible GPU (recommended for training)

📦 Installation

1. Clone this repository:

git clone <repository-url>
cd PythonProject

2. Create and activate a virtual environment:

python -m venv .venv
source .venv/bin/activate  # On Linux/Mac
# or
.venv\Scripts\activate  # On Windows

3. Install required packages:

pip install torch ultralytics

🚀 Usage

1. Single Image Detection

# Using pre-trained YOLOv10
predict_single_yolo("path/to/image.jpg")

# Using custom-trained model
predict_single_my_model("path/to/image.jpg")

2. Batch Image Processing

# Process all images in a folder
predict(input_folder="datasets/bicycles/test", output_folder="./results")

3. Video Detection

# Detect bicycles in a video file
detect_video("path/to/video.mp4")

4. Model Training

# Train a custom model from scratch
train()

📁 Project Structure

PythonProject/
├── yolo.py                 # Main script with all functions
├── yolov10n.pt            # Pre-trained YOLOv10 weights
├── yolo11n.pt             # Alternative YOLOv11 weights
├── datasets/
│   ├── bicycles-2000/     # Custom bicycle dataset
│   │   ├── train/         # Training images
│   │   ├── valid/         # Validation images
│   │   ├── test/          # Test images
│   │   └── data.yaml      # Dataset configuration
│   └── bicycles/          # Additional bicycle dataset
├── results/                # Output folder for image processing
├── video_results/          # Output folder for video processing
├── ultralytics/            # Ultralytics framework files
└── .venv/                  # Virtual environment

🎯 Dataset

The project uses a custom bicycle dataset with:

• Training set: 2000+ bicycle images
• Validation set: For model evaluation
• Test set: For inference testing
• Classes: 1 (Bicycle)
• Source: Roboflow dataset

🔧 Configuration

Training Parameters

• Epochs: 100
• Image Size: 320x320
• Batch Size: 8
• Model: YOLOv10n (nano version)

Model Paths

• Custom Model: ultralytics/runs/detect/my_custom_model_2000/weights/best.pt
• Pre-trained: yolov10n.pt

📊 Performance

The project includes both pre-trained YOLOv10 models and custom-trained models specifically optimized for bicycle detection. Custom models are trained from scratch on bicycle datasets for improved accuracy in bicycle detection scenarios.

🖥️ Hardware Requirements

• Minimum: CPU with 8GB RAM
• Recommended: NVIDIA GPU with 8GB+ VRAM
• Storage: 10GB+ free space for datasets and models

🚀 Quick Start

1. Check GPU availability:

python yolo.py

2. Run video detection (default):

# The script runs detect_video("bike.mp4") by default
python yolo.py

3. Train custom model:

# Uncomment train() in main section
python yolo.py

📝 Examples

The project includes several example images for testing:

• bicycle.jpg - Single bicycle
• bicycle_wbg.jpg - Bicycle with background
• bicycle_with_girl.jpg - Bicycle with person
• img.png, img_1.png, img_2.png - Various test images

🤝 Contributing

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Test thoroughly
5. Submit a pull request

📄 License

This project uses the CC BY 4.0 license for the dataset and is open source.

🙏 Acknowledgments

• Ultralytics for YOLOv10 implementation
• Roboflow for the bicycle dataset
• PyTorch team for the deep learning framework

📞 Support

For questions or issues, please open an issue on the project repository.

---

Note: This project is designed for research and educational purposes. Ensure you have proper permissions for any images or videos you process.