Description

Introduction

Vehicle detection is one of the autonomous car applications to detect nearby vehicles to avoid collisions. This application can also be used for traffic management based on traffic density. This project helps you to work with XML files which are a pre-trained model to detect the vehicles.

Abstract

In this project Nvidia Jetson Nano is used, USB camera is also interfaced with the Nano. Here cars.xml file is used. The algorithm with the model is in the XML file. This real-time project will be a base model to develop a vehicle detection system in traffic management or in an autonomous vehicle.

Existing system

In the existing system, video processing application to detect the vehicle is difficult to program using Matlab

Proposed system

In this proposed system, Computer Vision is used, which makes easy programming by loading cars.xml file, to detect the vehicle. The same thing becomes more efficient by choosing Nvidia Jetson Nano as the core.

Demo video

Connection description

USB camera is interfaced with the Nvidia Jetson Nano.

Project description

This project uses the Nvidia Jetson Nano, which has the GPU. So that video processing is much more efficient than other boards like Raspberry Pi. Cars.xml file which reduces the programming efforts which has the capability to classify the vehicle from any environment. This single project will help you to develop a different variety of applications.

Hardware Required

• Nvidia Jetson Nano
• 32GB SD card
• USB Camera

Software required

• OS for Nvidia nano
• SD card formatter
• Etcher

Result

This same program can also be run in Raspberry pi, but you will find the clear difference in the real-time video processing difference between Raspberry Pi and the Jetson Nano that Nano is better board for Real-time video-based application. Raspberry Pi can also be better, by integrating Intel Movidius Neural Compute Stick with the Raspberry Pi.