We are proposing a software allowing to detect, recognize and track traffic signs. This ongoing work is available
This work is attached to this research
Our contribution is threefold:
  • Proposition of a new naive detection based on color and shape retrieval.
  • Using the information of the detection, we propose to recognize traffic sign using machine learning methods.
  • In order to perform to accelerate the process of detection-recognition, we propose to introduce a tracking module information already extracted in the two previous stages.
This work has been developed by students of Ms in Computer Vision of UB and Guillaume Lemaître under supervision of Yohan Fougerolles.


I proposed some enhancements to the software GoldenCheetah. GoldenCheetah is an open-source cross platform software which allows to manage cycling data in order to train and have a rational approach to optimize the performances in cycling. A french article is making a description of this software [here].

Release & developer version can be found [here].