LBW Review System 🏏

The LBW (Leg Before Wicket) Review System is a computer vision-based project that analyzes cricket match videos to determine if a batsman is out or not based on LBW rules.

This system utilizes YOLOv8 for object detection and Flask for the backend to process cricket match videos.

---

🚀 Features

✅ Video Upload – Upload cricket match videos for analysis.
✅ Object Detection – Detects ball, batsman, stumps, and pitch using YOLOv8.
✅ LBW Decision System – Applies cricket LBW rules to determine if the batsman is out.
✅ User Interface – Simple web-based UI for easy interaction.

---

🛠 Installation

Prerequisites

• Python 3.8+
• pip (Python package manager)
• Git installed

1️⃣ Clone the Repository

git clone https://github.com/your-username/LBW-Review-System.git
cd LBW-Review-System

2️⃣ Install Dependencies

pip install -r backend/requirements.txt

3️⃣ Download YOLOv8 Model

• Download yolov8n.pt from the Ultralytics Repository.
• Place it in the models/ folder.

4️⃣ Prepare Dataset

• Store training data inside dataset/ folder.
• Annotate using LabelImg (GitHub).
• Modify dataset/dataset.yaml accordingly.

5️⃣ Train the Model (Optional)

python backend/train.py

6️⃣ Run the Backend Server

python backend/app.py

7️⃣ Open the Frontend

• Open frontend/index.html in a browser.
• Upload a cricket match video and get LBW results.

---

🎮 How to Use

1️⃣ Upload a cricket match video via the web interface.
2️⃣ System processes the video using YOLOv8 for object detection.
3️⃣ LBW logic is applied to detect if the batsman is Out or Not Out.
4️⃣ Results are displayed on the web interface.

---

📊 Dataset Structure

dataset/
├── images/
│   ├── train/            # Training images
│   └── val/              # Validation images
├── labels/
│   ├── train/            # YOLO format labels for training
│   └── val/              # YOLO format labels for validation
└── dataset.yaml          # Dataset configuration

• Use LabelImg to annotate images.
• Save annotations in YOLO format.

---

🧠 Model Training

1️⃣ Ensure dataset is properly structured.
2️⃣ Train model using:

python backend/train.py

3️⃣ Trained model is saved at runs/detect/train/weights/best.pt.
4️⃣ Move best.pt to the models/ folder for inference.

---

---

License 📜

This project is licensed under the MIT License - see the LICENSE file for details.