Cicada 3302

Copy and paste the image into an image editor (like Microsoft Paint) and use the fill tool to make the background a different color.

The hidden number $781386592$ will then appear!

The picture has a high contrast, so we have practically only the colors white and black. It is therefore interesting to consider the distribution of grayscale colors. For example, you can load the image in Python with the PIL library and convert it to a numpy array so that you can plot the statistics of the color values with matplotlib. Each pixel of the PNG file has a gray value between 0 (black) and 255 (white). In our case, the following statistic results:

$\begin{array}{r|ccccccc} \text{gray value} & 0 & 1 & 2 & \dots & 253 & 254 & 255 \\ \text{pixels} & 14844 & 313381 & 0 & \dots & 0 & 28746 & 3029 \end{array}$

It is noticeable that there are only the four gray levels 0, 1, 254 and 255 in the whole picture. In particular, most pixels have the value 1 or 254. This is by no means a random pattern, but indicates that here additional information has been encoded in the image. Since the gray values 0 and 1 or 254 and 255 are very close to each other, it is impossible to recognize the information with the naked eye. But if we create a new image in which we color all the pixels with the gray value 1 and 254 white and everthing else black, we get the following:

This picture is created by the following python code:

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16

import numpy as np
from PIL import Image

pic = Image.open('encodedpic.png')  # open image as an Image object
picBW = pic.convert('L')            # convert it to a grayscale image
picArray = np.array(picBW)          # create a two dimensional array from the grayscale image

for l in range(len(picArray)):                              # iterate over all pixels of the image 
    for p in range(len(picArray[l])):
        if picArray[l][p] == 0 or picArray[l][p] == 255:    # if an 'interestiong' pixel is found
            picArray[l][p] = 255                            # color it white
        else:
            picArray[l][p] = 0                              # otherwise color it black

result = Image.fromarray(picArray)  # convert the array back to an Image object 
result.save('hiddencode.png')       # and save it.