Proj S31 – KLASS Engineering – Mobile Vehicle Make Model Recognition

Introducing Proj S31 – KLASS Engineering – Mobile Vehicle Make Model Recognition

VMMRoom aims to aid enforcement agencies in identifying criminal activities associated with fraudulent licence plate mismatches in real-time. By combining the licence plate, vehicle make and model recognition systems, agencies can match vehicle licence plates to vehicular profiles accurately in real-time, thereby enabling timely investigations into any anomalies.

Team members

Lawrence Kevin Sagaya Anthony (CSD), Tan Zheng Yang (EPD), Lee Jing (CSD), Nigel Keng Yi Qian (EPD), Ning Zhi Yan (EPD)

Instructors:

  • Kwan Wei Lek

Writing Instructors:

  • Tay Zi Dong Patrick

BACKGROUND

Most cases of licence plate mismatches highlighted in the news are typically uncovered accidentally through the investigation of other crimes that may not be directly related to the fraudulent swapping of licence plates. Hence, there needs to be a way to help traffic enforcement identify such cases in a more intentional and timely manner.

How can we assist authorities in proactively detecting licence plate mismatches?

Automate detection process
Alert users on detection in real-time
Archive all detected activities for further investigation if needed

Introducing Our Product

VMMRoom

main
VMMRoom – a combined ANPR-VMMR system – aims to help law enforcement more actively identify fraudulent license plate mismatches.
In doing so, the system also helps law enforcement to identify more serious crimes.
An alert is displayed onto the in-vehicle screen to inform on-field traffic enforcement officers when a suspicious vehicle is detected. Officers in the command centre are also able to view information relating to the detected suspicious vehicle through the web user interface.
detection

Real-Time Detection

The system uses the KLASS-developed ANPR & VMMR models to predict licence plates & vehicle-make of road-going vehicles. Through various iterations, the system is able to perform real-time detection, achieving a processing time of 10 to 30 seconds.

Alerts through User Interface

Alerts of suspected vehicles are made known to ground officers via a user interface. It displays only the key information of the vehicles of interest to reduce distraction to ground officers. This includes the captured image frame, detected license plate, and instructions from the command centre.
in-car alert
dashboard page

Storage for Past Data

The website summarises the number of detected vehicles into categories. Command centre officers can view information of previously detected vehicles for further investigations when needed and update the status of these suspected vehicles.

Secure Holstering

The mechanical structure is designed to mount the system securely to the back of the front passenger seat. This ensures that the system is protected against shocks and vibrations experienced on moving vehicles. The modular design allows for easy deployment on various fast response vehicles.
holstered
car charger

Low Power Consumption

The system boasts a relatively low maximum power consumption of approximately 25 watts. This is within the range of rated power supplied by most car chargers – 20 watts to 40 watts. With an appropriately sized charger, the system can conveniently tap on the car battery to operate.

GNSS Positioning Retrieval

Upon detection of a suspected vehicle, officers will use the retrieved location coordinates to locate any suspicious activity. This allows for more resources to be directed there for investigations if necessary.

ENVISIONING A SAFER FUTURE

Our team hopes that with VMMRoom, we will be able to introduce another relevant tool for law enforcement to track down serious crimes & improve public safety.

Acknowledgements

Thank you to our Industry, Capstone, and Centre for Writing and Rhetoric (CWR) Mentors:
Dr Kwan Wei Lek, for his invaluable insights and technical assistance and loaning of equipment whenever possible.
Mr Tay Zi Dong, Patrick, for providing very detailed feedback on the communicative aspects of our Capstone deliverables.
Mr Ng Jun Wah, Ms Toh Si Hui, & Mr Lu Zheng Hao, from KLASS for working closely with us & providing product feedback.

In Collaboration With:

Menu

ornament-menu

Contact the Capstone Office :

+65 6499 4076

8 Somapah Road Singapore 487372

Please fill in your information below and feedback

Would you like to play the audio?

Contact the Capstone Office :

8 Somapah Road Singapore 487372

8 Somapah Road Singapore
487372

Welcome back!

Log in to your existing account.

Contact the Capstone Office :

+65 6499 4076

8 Somapah Road Singapore 487372

Welcome back!

Log in to your existing account.