Vehicle Tagging using Deep Learning

Project Title

Vehicle Tagging using Deep Learning

Project Area of Specialization Artificial IntelligenceProject Summary

Our goal is to perform real-time deep learning based object detection at security cameras, using a small, low-power setup made up of a Raspberry Pi and an Intel Movidius NCS.

Project Objectives

Features our solution will provide:

? Fully independent nodes/no single point of failure

? Autonomous vehicle recognition/tracking

? Central database only for synchronization hence

  low bandwidth requirement

Project Implementation Method

Our system will use a Raspberry pi along with an Intel Movidius NCS to perform real time analysis on video being fed to it via a camera. The system will also be connected online to enable communication between different nodes.

Benefits of the Project

Traffic surveillance is becoming an increasingly complex task due to the explosive growth in the number of vehicles on roads. It is no longer possible for humans alone to effectively manage such systems. We want to use deep learning and computer vision concepts to automate traffic surveillance.

Technical Details of Final Deliverable

The final deliverable will be a Raspberry pi/Intel Movidius combo connected to a camera. This device will be able to perform its tasks at the edge with minimal human intervention.

Final Deliverable of the Project HW/SW integrated systemType of Industry Transportation Technologies Artificial Intelligence(AI), Internet of Things (IoT), Cloud InfrastructureSustainable Development Goals Industry, Innovation and Infrastructure, Sustainable Cities and CommunitiesRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	47800
Intel Movidius NCS	Equipment	2	12600	25200
Raspberry Pi	Equipment	2	6300	12600
Camera	Equipment	2	4000	8000
SD card	Equipment	2	1000	2000