#Traffic Sign Recognition

##Writeup Template

###You can use this file as a template for your writeup if you want to submit it as a markdown file, but feel free to use some other method and submit a pdf if you prefer.

---

Build a Traffic Sign Recognition Project

The goals / steps of this project are the following:

• Load the data set (see below for links to the project data set)
• Explore, summarize and visualize the data set
• Design, train and test a model architecture
• Use the model to make predictions on new images
• Analyze the softmax probabilities of the new images
• Summarize the results with a written report

Rubric Points

###Here I will consider the rubric points individually and describe how I addressed each point in my implementation.

---

###Writeup / README

###Data Set Summary & Exploration

####1. Provide a basic summary of the data set. In the code, the analysis should be done using python, numpy and/or pandas methods rather than hardcoding results manually.

I used the pandas library to calculate summary statistics of the traffic signs data set:

• The size of training set is 34799
• The size of the validation set is 4410
• The size of test set is 12630
• The shape of a traffic sign image is (32,32,3)
• The number of unique classes/labels in the data set is 43

####2. Include an exploratory visualization of the dataset.

Here is an exploratory visualization of the data set. It shows a sample from each class and a label counts.

###Design and Test a Model Architecture

####1. Describe how you preprocessed the image data. What techniques were chosen and why did you choose these techniques? Consider including images showing the output of each preprocessing technique. Pre-processing refers to techniques such as converting to grayscale, normalization, etc. (OPTIONAL: As described in the "Stand Out Suggestions" part of the rubric, if you generated additional data for training, describe why you decided to generate additional data, how you generated the data, and provide example images of the additional data. Then describe the characteristics of the augmented training set like number of images in the set, number of images for each class, etc.)

As a first step, I normalized the image data so that it would have mean zero and equal variance.

As a second step, I decided to shuffle the data to get rid of bias.

####2. Describe what your final model architecture looks like including model type, layers, layer sizes, connectivity, etc.) Consider including a diagram and/or table describing the final model.

My final model consisted of the following layers:

Layer	Description
Input	32x32x3 RGB image
Convolution 5x5	1x1 stride, VALID padding, outputs 28x28x6
RELU	activation
Max pooling	2x2 stride, outputs 14x14x6
Convolution 5x5	1x1 stride, VALID padding, outputs 10x10x16
RELU	activation
Max pooling	2x2 stride, outputs 5x5x16
Fully connected	outputs 120
RELU	activation
DROPOUT	keep_prob = 0.5
Fully connected	outputs 84
RELU	activation
DROPOUT	keep_prob = 0.5
Fully connected	outputs 43

####3. Describe how you trained your model. The discussion can include the type of optimizer, the batch size, number of epochs and any hyperparameters such as learning rate.

To train the model, I used the Tensorflow CNN architecture with EPOCHS = 5, BATCH_SIZE = 256 and learning rate = 0.0009. 5 epochs was enough to get a validation accuracy over 95% and after a few epochs accuracy didn't really improve significantly.

####4. Describe the approach taken for finding a solution and getting the validation set accuracy to be at least 0.93. Include in the discussion the results on the training, validation and test sets and where in the code these were calculated. Your approach may have been an iterative process, in which case, outline the steps you took to get to the final solution and why you chose those steps. Perhaps your solution involved an already well known implementation or architecture. In this case, discuss why you think the architecture is suitable for the current problem.

My final model results were:

• train set accuracy of 0.999
• validation set accuracy of 0.961
• test set accuracy of 0.945

To avoid overfitting I included dropout layers. But since I didn't create an augmented dataset, I think the train set accuracy is too high and this is an indication of some overfitting. Nevertheless, the high accuracy in validation and test sets was satisfying.

And to increase validation accuracy, I reduced the learning rate to 0.0009 after some trial and error.

For this problem, convolutional layers worked well since we are dealing with images. My architecture is the LeNet architecture.

###Test a Model on New Images

####1. Choose five German traffic signs found on the web and provide them in the report. For each image, discuss what quality or qualities might be difficult to classify.

Five German traffic signs that I found on the web are in the folder test_images.

####2. Discuss the model's predictions on these new traffic signs and compare the results to predicting on the test set. At a minimum, discuss what the predictions were, the accuracy on these new predictions, and compare the accuracy to the accuracy on the test set (OPTIONAL: Discuss the results in more detail as described in the "Stand Out Suggestions" part of the rubric).

Here are the results of the prediction:

Image	Prediction
Right-of-way	Right-of-way
No entry	No entry
Speed 30 km/h	Speed 30 km/h
Stop Sign	Stop sign
Speed 20 km/h	Speed 80 km/h

The model was able to correctly guess 4 of the 5 traffic signs, which gives an accuracy of 80%. I feel like this accuracy would be higher if I used more than five images and more classes.

####3. Describe how certain the model is when predicting on each of the five new images by looking at the softmax probabilities for each prediction. Provide the top 5 softmax probabilities for each image along with the sign type of each probability. (OPTIONAL: as described in the "Stand Out Suggestions" part of the rubric, visualizations can also be provided such as bar charts)

Except the last image, which is Speed 20 km/h, the model was correctly predicting the labels with 1.00 probability.

For the last image, the model is relatively sure that this is a Speed 80 km/h.