MyGarage is a university-based AI parking system that uses machine learning and sensors to automate parking spot management, detect license plates, and control access to parking gates.
Parking on university campuses is normally slow, manual, and frustrating. MyGarage improves the process by using AI (YOLOv8 + OCR) and Arduino hardware to detect cars, read plate numbers, and open gates automatically.
The goal is to make parking faster, smarter, reduce errors, and improve student/employee experience.
- Arduino Uno R3
- Ultrasonic Sensors
- 9g Servo Motor (gate opening)
- Breadboard + Wires
- Python (YOLOv8 for plate detection)
- EasyOCR (plate text recognition)
- OpenCV (image processing)
- Flutter (Mobile UI)
- SQLite (Local DB)
- Roboflow (Dataset Annotation)
-
2000 images collected from different angles
- YOLOv8 detects the car plate and plate ID
- Recognition accuracy: 54% – 71.5%
- Detection success rate: 96.5%
The system automatically:
- Detects car approaching
- Detects plate area
- Reads plate ID & number
- Can open gate automatically (servo motor)
- Add backend CRUD API
- Improve OCR recognition accuracy
- Upgrade sensors + wiring
- Handle more students/employees at scale
- Privacy & data protection enhancements
- Yara Hakij
- Lana Abdulhaq
- Anas Salman
Supervisor: Prof. Mohammad I. Daoud
B.Sc. Computer Engineering – German Jordanian University, 2024