Hi, in this lesson, you're going to learn how to use functions to implement what is called steganography. Hiding data in an image or other digital artifact like an audio file, a software program, or any file of zeros and ones. Because steganography is a bit larger of a task than previous problems you have solved. It's a great idea to break the code down into functions. As you progress through this lesson, you'll see some examples where pieces of code are abstracted out into separate functions. To simplify the entire solution and to avoid repetition of code. The idea behind steganography is to take an image, like this image of sprinter Usain Bolt. And hide data like another image inside it. Altering the numerical values of the pixels of Usain Bolt to encode the hidden image. The key to steganography is to hide the data in such a way that nobody can easily tell you've altered the original image. This image on the right is the result of hiding a secret message in the first image. Can you tell that it has been altered? If you look really closely, you can see some differences in the shading of the background. However, if you were to just look at the second image by itself, there is nothing suspicious. It just looks like a picture of Usain Bolt. This idea has been around for a long time, predating computers. Sending an undetected message has been important throughout history. One important modern use is to avoid censorship imposed by oppressive governments. You can hide any digital information in an image. You could, for example, hide a text or HTML file in an image. However, that would take a bit more math and a deeper understanding of how everything is a number. To keep things simple at first, you'll work with hiding an image in another image of the same size. When you are finished with this lesson, you'll be able to find hidden meaning in the universe. That is, we're going to walk you through not only the concepts, but also the codes to do steganographic hiding. Then you will write the code to extract the hidden message from an image. For example, you could take this picture of a galaxy and extract the message that we hid inside of it. Okay, so how do you actually do this? You already know that pixels have red, green and blue components which numerically represent the color. Is there a big difference between a red value of 240 and a red value of 255? They are numerically different, but if you look at them, the two of them are pretty similar. The fact that you can't easily tell the difference when you change the numerical value slightly is the key to hiding data in an image. You can store the hidden data in the least significant digits of the color value. And not see much change in the resulting color. The least significant digits are like the ones and tens places in a three digit number. So you could hide a 15 in 240 by changing 240 to 255. To do this, you're going to need a bit of math. Don't worry, it's just multiplication, division and addition. You just have to put them together the right way. To understand how to do this math, we're going to start with decimal. The base ten number system you use everyday. We'll explain the concepts in base ten. Then learn about binary, the base two number system that the computer uses to store numbers. All the principles will be the same in base two as they are in base ten. You will just need to use powers of two instead of powers of ten. To see this idea in the familiar base ten. Suppose that the values for the red, green and blue went from 0 to 9999 instead of 0 to 255. This gives each component of the color 4 decimal digits. Now, suppose you want to hide this red pixel with the RGB values shown. Red = 8274, Green = 0 and Blue = 1098. In this blue pixel with RGB values of Red = 3568, Green = 5686 and Blue = 7450. We'll put the result here on the right. For the red, you want to take the most significant two digits of the pixel you want to hide the data in, which is the blue pixel. And use them as the most significant two digits of the result. Then you want to take the most significant two digits of the pixel you want to hide, the red pixel. And use them as the least significant two digits of the result. Notice that 3582 is pretty similar to 3568. It will look almost the same. But you've changed it slightly so that it stores the secret information in its least significant digits. Now, you do the same thing for green. Taking the two most significant digits from this blue pixel. And combining them with the most significant two digits from this red pixel. Again, 5600 is pretty similar to 5686. Now, you do the same thing for the blue component. Combining the most significant digits from the two pixels' blue components. The resulting number, 7410 is again quite similar to the original 7450. If you look at the resulting color of this pixel, it's pretty hard to tell the difference between it and the original blue pixel. But as you'll see, we've hidden a red pixel inside it. Now that the information is hidden, how would you extract the secret? You know that you want the least significant two digits of this pixel's red to be the most significant two digits of the hidden and soon to be extracted pixel's red. So we want 82 to be the most significant digits of the R or red value. But what should the least significant digits be? It doesn't really matter too much, so we'll just pick 0. Then, you do the same thing for green, and the same thing for blue. If you look at the resulting color, it is this shade of red. That shade of red is pretty close to the original color that we wanted to hide. Even though you don't get the exact color back out, the extracted image will look pretty similar. So now you know the big idea. Steganography is hiding data in other data. In particular, you're going to do this with hiding one image in another. Now you understand the basic math that is involved in it, taking digits from each number and combining them. However, to implement this in code, you are going to need to learn a little bit about binary. The base two number system the computer uses. Which is why the color values go from 0 to 255 instead of the 0 to 9999 that we just worked with.
