Traffic-sign recognition: Difference between revisions

'''Traffic-sign recognition''' ('''TSR''') is a technology by which a vehicle is able to recognize the [[traffic sign]]s put on the road e.g. "speed limit" or "children" or "turn ahead". This is part of the features collectively called [[Advanced driver-assistance systems|ADAS]]. The technology is being developed by a variety of automotive suppliers to improve the safety of vehicles. It uses image processing techniques to detect the traffic signs. The detection methods can be generally divided into color based, shape based and learning based methods.

Traffic-sign recognition first appeared, in the form of speed limit sign recognition, in 2008 for the 2009 [[Vauxhall Insignia]].<ref>{{cite news|date=18 June 2008|title=Vauxhall Insignia to read speed limit signs|url=https://www.businesscar.co.uk/news/2008/vauxhall-insignia-to-read-speed-limit-signs|work=Business Car|access-date=2 April 2019}}</ref> Later in 2009 they appeared on the new [[BMW F01|BMW 7 Series]], and the following year on the [[Mercedes-Benz S-Class]]. At that time, these systems only detected the round speed limit signs found all across Europe (e.g.<ref>{{cite journalbook|last=Eichner|first=M.|author2=Breckon, T. |title=2008 IEEE Intelligent Vehicles Symposium |chapter=Integrated speed limit detection and recognition from real-time video|journal=IEEE International Intelligent Vehicles Symposium|year=2008|pages=626–631|doi=10.1109/IVS.2008.4621285|chapter-url=http://community.dur.ac.uk/toby.breckon/publications/papers/eichner08speedlimit_a.pdf|isbn=978-1-4244-2568-6|s2cid=12477544}}</ref>).

Second-generation systems can also detect overtaking restrictions. It was introduced in 2008 in the [[Opel Insignia]],<ref>{{cite web |url=http://www.autobild.de/artikel/assistenzsystem-von-opel-722126.html|title=Assistenzsystem von Opel – Das magische Auge|date=18 June 2008 |accessdate=17 December 2010}}</ref> later followed by the [[Opel Astra]] and the [[Saab 9-5]]. This technology is also available on the 2011 [[Volkswagen Phaeton]]<ref>{{cite web |url=https://www.volkswagen-media-services.com/medias_publish/ms/content/en/pressemitteilungen/2010/04/22/phaeton_debuts_with.standard.gid-oeffentlichkeit.html|title=Phaeton debuts with new design and new technologies|accessdate=22 April 2010|archive-date=20 July 2011|archive-url=https://web.archive.org/web/20110720152841/https://www.volkswagen-media-services.com/medias_publish/ms/content/en/pressemitteilungen/2010/04/22/phaeton_debuts_with.standard.gid-oeffentlichkeit.html|url-status=dead}}</ref> and, since 2012, in the [[Volvo S80]], [[Volvo V70|V70]], XC70, [[Volvo XC60|XC60]], [[Volvo S60|S60]], [[Volvo V60|V60]] and [[Volvo V40 (2012–2019)|V40]], as a technology called ''Road Sign Information''.<ref>{{cite web |url=https://www.media.volvocars.com/global/enhanced/en-gb/Media/Preview.aspx?mediaid=47896|title=Road Sign Information|accessdate=19 February 2013}}</ref> They are not able to recognize city limit signs, which in most European countries are associated with speed limits, as they are too similar to direction signs.

== WorkingImplementation ==

Modern traffic-sign recognition systems are being developed using convolutional neural networks, mainly driven by the requirements of autonomous vehicles and [[self-driving carscar]]s. In these scenarios, the detection system needs to identify a variety of traffic signs and not just speed limits. This is where the [[Vienna Convention on Road Signs and Signals]] comes to help. A convolutional neural network can be trained to take in these predefined traffic signs and 'learn' using [[Deep Learning]] techniques.

The neural net in turn uses [[Image Processing]] and [[Computer Vision]] to train the network with its potential outcomes. The trained neural net can then be used in real time to detect new traffic signs in real time. Self driving car companies including [[Waymo]] and [[Uber]] are generating and outsourcing traffic-sign data sets along with map and navigation companies like [[TomTom|Tom Tom]].<ref>{{cite web |url=http://www.popularmechanics.com/cars/a21609/here-maps-future-of-self-driving-cars/|title=Whoever Owns the Maps Owns the Future of Self-Driving Cars|date=July 2016 }}</ref> Advanced [[computer vision]] and neural network techniques make this goal highly efficient and achievable in real time.

Deep learning can be incorporated into traffic-sign detection. Polygonal approximation of digital curves using [[Ramer–Douglas–Peucker algorithm]] can be used to detect the shape of the sign boards and methods like [[Support Vector Machines]] and Byte-MCT with an [[AdaBoost]] classifier has been used in one of the methods to detect traffic signs.<ref>{{cite journal |title=Lim K, Hong Y, Choi Y, Byun H (2017) Real-time traffic-sign recognition based on a general-purpose GPU and deep learning. PLoS ONE 12(3): e0173317|year=2017|doi=10.1371/journal.pone.0173317|pmid=28264011|doi-access=free|last1=Lim|first1=K.|last2=Hong|first2=Y.|last3=Choi|first3=Y.|last4=Byun|first4=H.|journal=PLOS ONE|volume=12|issue=3|pages=e0173317|pmc=5338798}}</ref>

=== Groups =Usage==