This repository stores some of the source codes developed in the computer vision course (Visao Computacional @ UnB) during the 1/2017 semester.
This project has as goal a software system to tracking cars on a road obtaining information about their position and the number of cars on a road.
Antoni B. Chan and Nuno Vasconcelos, In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), San Diego, Jun 2005.
http://visal.cs.cityu.edu.hk/downloads/ http://www.ipf.kit.edu/downloads_data_set_AIS_vehicle_tracking.php http://www.tft.lth.se/en/research/video-analysis/co-operation/data-exchange/
This project was developed and tested in a GNU-Linux computer. To compile it's necessary to have OpenCV 2.4 installed and CMake >= 2.8.
Open terminal: To compile: $ cmake . $ make
To run: $ ./TrafficSurveillance
To run executing a specific video $ ./TrafficSurveillance <name_of_the_video>
To change the algorithms being used in this version it is necessary to change the source code and recompile
- To change the detecion method: Change the value of the define DETECTOR_USED at line 62 of the file main.cpp
- To change the tracking method: Change the value of the define TRACKER_USED at line 63 of the file main.cpp