Laptop Protection using YOLOv9
This project focuses on building a real-time laptop monitoring and protection system using YOLOv9 object detection. The goal is to detect laptops accurately in different environments and improve security in places like colleges, libraries, offices, and workspaces.
Dataset & Annotation
We created a custom dataset using images and video frames collected from different angles, lighting conditions, and backgrounds. The dataset was annotated with bounding boxes for accurate object detection training.
Preprocessing
- Image resizing
- Dataset splitting
- Data augmentation
- Frame extraction from videos
Model Training
The model was trained using YOLOv9s.pt on a custom dataset for real-time object detection.
Training Highlights
- Custom trained YOLOv9 model
- Precision, Recall & mAP monitoring
- Improved accuracy through retraining and augmentation
During training, challenges like low-light detection and partially visible laptops were addressed by improving dataset quality and annotations.
Deployment & Results
The trained model was deployed on the YOLOvX platform for real-time inference and testing.
Results
- Real-time laptop detection achieved
- Smooth bounding box visualization
- Improved detection accuracy in live testing
Conclusion
This project provided hands-on experience in:
- Dataset creation
- Annotation
- Model training & validation
- Real-time deployment using YOLOv9
It helped us understand the complete workflow of Vision AI and object detection systems in practical applications.
Demo Video:https://drive.google.com/file/d/1Vut7QCOBFSl-YCnISoEpp5kxQ3QSeSgn/view?usp=drivesdk