Traffic Sign Recognizer

The traffic signs perform a priceless job by acting as guidelines for both drivers and pedestrians on the road. While the traffic signs are supposed to be understood by human vision, it is interesting to if the same phenomenon can be emulated using computer vision. In this demonstration, we attempt to address a basic problem, in which we check the possibility of recognizing the traffic signs via the concepts of computer vision.

Instructions

  • Choose an appropriate traffic sign image from your computer by selecting an image file from the "Choose File" option given below.
  • Please make sure that the chosen image approximately resembles a 1:1 ratio between the width and height of the image (E.g.: 30x30, 60x58, 150x150, etc.)
  • Some samples of the accepted images, and the list of traffic sign classes accepted by the application, are given below for your reference.
  • Once you choose an image, a preview of the chosen image will be shown under Image Preview
  • Click on Classify
  • The sign recognizer will make a classification based on the image submitted by you, and the image predicted by the classifier will be shown under Predicted Image as its meta image equivalent
  • Under Result, more information regarding the classification will be shown. Class Label denotes the numerical label which we have assigned for each class of traffic signs. Sign Name is a reader-friendly annotation for the sign, whereas the Confidence Level demonstrates the probability of the predicted image being correct.
  • Enjoy! 😊

An error occurred while processing the image. Please try again or try the classification using a different image.

Image Preview
Your Image
Predicted Image
Predicted Image
Result
Class Label
N/A
Sign Name
N/A
Confidence Level
N/A