Vehicle Detection, Tracking and Speed Estimation

Background: Predicting high speed is very important for preventing potential accidents. Approximately $1,3$

million people die each year because of road traffic crashes according to the World Health Organization.

Task: Develop a speed tracking system that detects cars on a highway and estimates their speed over time $($i$.$e$.,$

km$/$h or miles$/$h$).$ If a car exceeds a predefined speed limit$,$ the system should trigger an alert about that car. This

could be drawing a red box around the car, for example. There are many algorithms available that can be used for

detecting and tracking cars. However, the quality of each of these algorithms in terms of accuracy may vary. A

comparison of three state of the art approaches and how their accuracy affects the speed prediction should be

addressed in the report. The approach used for deriving the speed of a vehicle should also be addressed in the

report. Any programming language$($s$),$ computer vision framework$($s$)$ and open$-$source libraries can be used for

the development of the speed tracking system. At the end of this evaluation, capture the speed tracking system

on a video sequence, the result of which should be provided as a link in the report. You can upload the video on

YouTube, OneDrive or any other secure cloud provider. Make the video available only if the link to the video is

provided. Develop a $10$ page report for this project.