In this study we have successfully identified an area of application for public video streams, which will produce great benefits for the community in general and for the industry as well. We have carried out a survey that can be used as a starting point for other studies in the field or related areas. By identifying the relevant issues in using public video streams, we were able to devise a methodological approach to start digging a bit deeper into the development of a platform to aid traffic flow characterization. Nonetheless, a similar methodological approach may be applied to other analogous problems.After the initial survey which gave us a general insight into this area and its main issues, a testing application has been designed and iteratively developed. The next steps include the improvement of the results achieved so far. This will be performed in two basic ways. Firstly, we intend to test with different parameter values, tuning the configuration of the implemented algorithms. Secondly, we plan to test with alternative algorithms and extend the ones already developed.For the latter case, AI-based techniques will play a major role. Additionally to analyzing traffic on an aggregate basis, individual vehicle detection and tracking are also features to be developed. Inarguably, this will pose great challenges and will bring about tricky issues to overcome when public video streams are to be used. Finally, a comprehensive and thorough calibration and validation methodology must be devised so as derived traffic flow information can be used and integrated intoother decision support systems.