Authors :
Anju C P; Andria Joy; Haritha Ashok; Joseph Ronald Pious; Livya George
Volume/Issue :
Volume 5 - 2020, Issue 7 - July
Google Scholar :
http://bitly.ws/9nMw
Scribd :
https://bit.ly/2FdunBz
DOI :
10.38124/IJISRT20JUL787
Abstract :
As placement of traffic sign board do not
follow any international standard, it may be difficultfor
non-local residents to recognize and infer the signs
easily. So, this project mainly focuses ondemonstrating
a system that can help facilitate this inconvenience. This
can be achieved byinterpreting the traffic sign as a voice
note in the user’s preferred language. Therefore, the
wholeprocess involves detecting the traffic sign,
detecting textual data if any with the help of
availabledatasets and then processing it into an audio as
the output to the user in his/her preferred language.The
proposed system not only tackles the above-mentioned
problem, but also to an extent ensuressafer driving by
reducing accidents through conveying the traffic signs
properly. The techniques usedto implement the system
include digital image processing, natural language
processing and machinelearning concepts. The
implementation of the system includesthree major steps
which are detection of traffic sign from a captured
traffic scene, classification of traffic signs and finally
conversion of classified traffic signs to audio message.
As placement of traffic sign board do not
follow any international standard, it may be difficultfor
non-local residents to recognize and infer the signs
easily. So, this project mainly focuses ondemonstrating
a system that can help facilitate this inconvenience. This
can be achieved byinterpreting the traffic sign as a voice
note in the user’s preferred language. Therefore, the
wholeprocess involves detecting the traffic sign,
detecting textual data if any with the help of
availabledatasets and then processing it into an audio as
the output to the user in his/her preferred language.The
proposed system not only tackles the above-mentioned
problem, but also to an extent ensuressafer driving by
reducing accidents through conveying the traffic signs
properly. The techniques usedto implement the system
include digital image processing, natural language
processing and machinelearning concepts. The
implementation of the system includesthree major steps
which are detection of traffic sign from a captured
traffic scene, classification of traffic signs and finally
conversion of classified traffic signs to audio message.
