In this project, i have used Convolutional Neural Network to build, train and test a traffic sign classification model
I have bhuild this model this mode using Tensorflow and keras. It is a multiclass classification problem. This model can be used to make smarter cars
- Finding data on kaggle and loading into colab. Download Dataset Size 314MB
- Preprocessing the image and visualizing them.
- Finding out mean of the dimenional and resizing all images accordingly.
- Converting the image into a numpy array and normalize them.
- Checking class imbalance.
- Splitting the data and performing ONE-HOT encoding
- Creating the model architcture, compiling the model and then fitting it.
- Plotting the accuracy and loss against each epoch.
- Preprocessing the test data and make prediction on it.
- Visualizing the original and predictied labels for the test image.