The design of vehicle classification system on low cost embedded board

Detection and classification of moving vehicle in a real time is relatively a challenging task, so in this project a vision based supervised system is presented which will detect vehicle through camera and classify them in three classes i.e. motorbike, bicycle, car. Our classifier model is based on

2025-06-28 16:36:18 - Adil Khan

Project Title

The design of vehicle classification system on low cost embedded board

Project Area of Specialization Artificial IntelligenceProject Summary

Detection and classification of moving vehicle in a real time is relatively a challenging task, so in this project a vision based supervised system is presented which will detect vehicle through camera and classify them in three classes i.e. motorbike, bicycle, car. Our classifier model is based on YOLO (You Only Look Once) algorithm and implemented on single embedded board which will increase the feasibility to use this system in real time. We are also working to  increase our system's computional power and improve efficiency &   accuracy through using Intel movidius NCS.

Project Objectives

1) Design and develop a system that works in real time and automatically classify vehicle type (Car, Bike, Truck).

2) Design a system that works on low computational power and low cost embedded board computers such as Raspberry Pi.

3) Reduce Computional complexity and improve accuracy by using Intel Movidius Neural compute stick.

Project Implementation Method

In this project, a system  is designed to locate on classify the vehicle type. So first of all the YOLO (you only look once) algorithm is designed using python language in three classes i.e. Car, Bicycle, and Motorbike. then this algorithm is shifted on Raspberry Pi board and tested in real time using web camera. 

Benefits of the Project

1) control traffic flow

2)  Automated toll plaza

3)  Automated parking area

4)  Road safety  

5)  Enhance monitoring capabilities

6)  Automobile companies for survey purpose

Technical Details of Final Deliverable

Finally, we developed a hardware system which includes a Raspberry Pi and a web camera attached.(Note: Intel Movidius NCS can be part of this project  in order to increase accuracy but we are still working on it.)

Final Deliverable of the Project Hardware SystemType of Industry Transportation , Security Technologies Artificial Intelligence(AI)Sustainable Development Goals Industry, Innovation and InfrastructureRequired Resources
Item Name Type No. of Units Per Unit Cost (in Rs) Total (in Rs)
Total in (Rs) 22400
Raspberry Pi 3b+ Equipment163006300
Web Camera Equipment1800800
Intel Movidius NCS Equipment11530015300

More Posts