Detailed description about the project is explained clearly in report(pdf) file.
This repository consists of a cascade classifier file(cars.xml) and python file(speed_detect.py) to use these cascade classifiers to detect cars. frames_per_sec.py file is used to calculate the number of frames per sec in a given video which is used in calculating speed. plate.py file is used to extract information related to the car by querying from Roads and Martitime services website using API.
The image database can be downloaded from here : http://cogcomp.cs.illinois.edu/Data/Car/
The explanation of code is done with comments.
Download the cascade file and the python file.
You are supposed to have several python libraries mentioned in the code installed in your system to get the code running.
Keep all the files in a single folder location.
Compile the speed_detect.py program in a usual OpenCV compilation way.
Run the code : example : ./speed_detect.py