diff --git a/README.md b/README.md index c49069c..d955084 100644 --- a/README.md +++ b/README.md @@ -1,8 +1,8 @@ -# 🚀 YOLOv8 Space Station Object Detection +# YOLOv8 Space Station Object Detection A complete object detection pipeline using YOLOv8 to detect **Toolbox**, **Oxygen Tank**, and **Fire Extinguisher** in space station environments using synthetic data from Duality AI Falcon platform. -## 🎯 Project Goals +## Project Goals - Achieve **≥90% mAP@0.5** accuracy - Robust detection under varying lighting conditions @@ -10,7 +10,7 @@ A complete object detection pipeline using YOLOv8 to detect **Toolbox**, **Oxyge - Real-time inference capabilities - Comprehensive evaluation and visualization -## 📋 Table of Contents +## Table of Contents - [Features](#-features) - [Installation](#-installation) @@ -22,9 +22,8 @@ A complete object detection pipeline using YOLOv8 to detect **Toolbox**, **Oxyge - [Configuration](#-configuration) - [Troubleshooting](#-troubleshooting) -## ✨ Features - -### 🎯 Core Functionality +## Features +### Core Functionality - **YOLOv8 Training Pipeline** with hyperparameter optimization - **Data Augmentation** (mosaic, HSV, flip, rotation, scaling) - **Comprehensive Evaluation** (mAP@0.5, confusion matrix, per-class metrics) @@ -32,21 +31,21 @@ A complete object detection pipeline using YOLOv8 to detect **Toolbox**, **Oxyge - **Real-time Web Application** with Streamlit - **Model Optimization Strategies** for accuracy improvement -### 📊 Analytics & Visualization +### Analytics & Visualization - Training curves and loss plots - Confusion matrix visualization - Per-class performance metrics - Dataset distribution analysis - Failure case documentation -### 🔧 Technical Features +### Technical Features - Modular and clean code architecture - Comprehensive logging and error handling - GPU/CPU support with automatic device detection - Configurable hyperparameters via YAML - Model checkpointing and resume training -## 🚀 Installation +## Installation ### Prerequisites - Python 3.8+ @@ -91,7 +90,7 @@ mkdir -p runs/{train,val} mkdir -p logs models results ``` -## 📁 Dataset Preparation +## Dataset Preparation ### Dataset Structure ``` @@ -139,7 +138,7 @@ python data_utils.py --action split_dataset --images-dir raw_images --labels-dir python data_utils.py --action visualize --images-dir dataset/train/images --labels-dir dataset/train/labels ``` -## 🎯 Training +## Training ### Basic Training ```bash @@ -173,7 +172,7 @@ Edit `config.yaml` to customize: - **Training summary**: `training_summary.txt` - **Logs**: `training.log` -## 📊 Evaluation +## Evaluation ### Model Evaluation ```bash @@ -199,7 +198,7 @@ python predict.py --model runs/train/yolov8_training/weights/best.pt --images da - **Failure cases**: `failure_cases.json` - **Optimization recommendations**: `optimization_recommendations.txt` -## 🌐 Web Application +## Web Application ### Launch Streamlit App ```bash @@ -219,7 +218,7 @@ streamlit run app.py 4. View detection results and statistics 5. Analyze performance in the analytics tab -## 📁 Project Structure +## Project Structure ``` space-station-object-detection/ @@ -243,7 +242,7 @@ space-station-object-detection/ └── results/ # Evaluation results ``` -## ⚙️ Configuration +## Configuration ### Key Configuration Parameters @@ -280,7 +279,7 @@ names: 2: Fire Extinguisher ``` -## 🎯 Model Optimization Strategies +## Model Optimization Strategies ### For Achieving ≥90% mAP@0.5 @@ -370,7 +369,7 @@ python data_utils.py --action analyze --dataset-dir dataset | YOLOv8m | 90-94% | 25.9 | 25.9 | | YOLOv8l | 92-96% | 43.7 | 43.7 | -## 🤝 Contributing +## Contributing 1. Fork the repository 2. Create a feature branch @@ -378,18 +377,18 @@ python data_utils.py --action analyze --dataset-dir dataset 4. Add tests if applicable 5. Submit a pull request -## 📄 License +## License This project is licensed under the MIT License - see the LICENSE file for details. -## 🙏 Acknowledgments +## Acknowledgments - **Ultralytics** for YOLOv8 implementation - **Duality AI** for Falcon synthetic data platform - **Streamlit** for web application framework - **OpenCV** for computer vision utilities -## 📞 Support +## Support For questions and support: - Create an issue on GitHub @@ -398,4 +397,5 @@ For questions and support: --- -**Happy detecting! 🚀🔍** \ No newline at end of file + +**Happy detecting!**