Skip to content


Subversion checkout URL

You can clone with HTTPS or Subversion.

Download ZIP
Track RECOGnition for Mobile Robots. Created at the University of São Paulo, Brazil.
C++ Shell
branch: master

Fetching latest commit…

Cannot retrieve the latest commit at this time

Failed to load latest commit information.

Track RECOGnition for Mobile Robots

This program is part of a project based on computer vision, artificial intelligence and robotics. Our objective is to find tracks in the visible path to mantain the vehicle securely between them.

This instance of our work is a single image treatment program that accepts as input an image file (JPG, TIFF, PNG, etc) and returns the lines corresponding to tracks. It uses OpenCV, an open source library for computer vision version 2.4+.


You need the OpenCV library installed in your computer to compile this program. Read out the OpenCV's documentation and wiki for more info.

To compile it:


I assume that the directory where the opencv header file (opencv2.hpp) is located is /usr/local/include/opencv2/. If it's different, you should change and export the environment variable INCLUDE_DIR. For example:

export INCLUDE_DIR=/path/to/opencv2/header

It should compile without errors if your instalation of OpenCV (2.4+) is correct. Open an issue on the issue tracker in case of problems.

The /samples folder contains some images to explore the possible outcomes of the program. The way to call this program is fairly straightforward:

./trecog [path/to/filename]

If you don't provide an argument, the program will use /samples/img1.jpg and, if it doesn't exist, it will generate an error.


Copyright (c) 2011-2013 Carlos Agarie


See LICENSE.txt for details.

Something went wrong with that request. Please try again.