Traffic sign identification

In this case study, we will understand the way in which we can classify a signal into one of the 43 possible classes.

Getting ready

For this exercise, we will adopt the following strategy:

1. Download the dataset that contains all possible traffic signs
2. Perform histogram normalization on top of input images:
   • Certain images are taken in broad day light, while others might be taken in twilight
   • Different lighting conditions result in a variation in pixel values, depending on the lighting condition at which the picture is taken
   • Histogram normalization performs normalization on pixel values so that they all have a similar distribution
3. Scale the input images
4. Build, compile, and fit a model to reduce...