27. Day and Night Classification

Day and Night Classification

You’re on your way to being able to build classifier! You know how to analyze color and brightness in a given image, and that skill alone can help you distinguish between different traffic signs.

Now, I’m going to give you a classification challenge. What if I asked you to classify two types of images, taken during the day or at night when the sun has set, and I want you to separate these images into two classes: day or night.

This is actually an important classification challenge for self-driving cars. These cars need to know what kind of conditions they are driving in, so that they can safely navigate a road at any time of day, and so that they can still recognize other vehicles and surrounding objects whether it is dark or light outside!

We’ll walk through each classification step together, but what do you think would be the first step in creating a classification model for day and night images?

Before you can classify any set of images, you have to look at them! Visualizing the image data you’re working with is the first step in identifying any patterns in image data and being able to make predictions about the data!

So, we’ll first load in this image data and learn a bit about the images we’ll be working with.