SmartBridge: Automated Crack Detection & Analysis

Overview

SmartBridge is an AI-powered bridge crack detection system utilizing a customized YOLO (You Only Look Once) model to enhance crack detection accuracy. This project enables real-time and automated structural health monitoring of bridges, reducing manual inspection efforts and improving infrastructure safety.

Features

• Customized YOLO Model: Fine-tuned for bridge crack detection
• Real-Time Processing: Efficient and fast crack identification
• High Accuracy: Optimized for minimal false positives and false negatives
• Easy Integration: Can be deployed on drones, robots, or handheld devices
• Data Logging & Reporting: Generates reports for maintenance planning

Installation

Prerequisites

Ensure you have the following installed:

• Python 3.8+
• PyTorch
• OpenCV
• TensorFlow (if needed)
• YOLO Framework

Clone the Repository

git clone https://github.com/Dr-irshad/SmartBridge-Automated-Crack-Detection-Analysis.git
cd SmartBridge-Automated-Crack-Detection-Analysis

Install Dependencies

pip install -r requirements.txt

Dataset Preparation

1. Collect bridge crack images and label them using tools like LabelImg.
2. Organize dataset:

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

3. Convert annotations to YOLO format.

Training the Model

To train the customized YOLO model:

python src/concrete_train.py 

Deployment

For real-time deployment, use:

python src/live_detect.py --weights best.pt --source 0  # 0 for webcam, video path for pre-recorded footage

Results & Visualization

• Detected cracks will be marked with bounding boxes.
• Reports can be exported in JSON or CSV format.

Contributions

We welcome contributions! To contribute:

1. Fork the repo
2. Create a feature branch (git checkout -b feature-name)
3. Commit changes (git commit -m 'Add feature')
4. Push to branch (git push origin feature-name)
5. Create a Pull Request

License

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

Contact

For queries, contact iikhaan@yahoo.com or create an issue in the repository.