Generates an image that consists of red, blue and green pixels randomly generated by IBM's quantum computer. The probabilities of the colours can be varied to approximate the desired shade.
This code was written in an attempt to build a logo for the quantum chessboard.
Generates an approximation to the colour you want using randomly generated red/blue/green dots generated with the probability of the ratio of the colour in its hex code. The generation is truly random as it is done using a quantum random number generator run on the IBM Quantum Experience. This makes this code a very basic application of a QRNG.
This is how it works:
Let the square I want to fill with a certain colour with an rgb value of (x,y,z) be divided into 40,000 small squares. Let's call them dots. The ratio of red, green and blue dots in the image should be x:y:z. Each of the 40,000 numbers assigns a certain colour- red, green or blue randomly to the respective dot, generated with the corresponding probability, which is set by the quantum circuit. Thus, the average ratio across the whole image is approximately x:y:z and matches the rgb value of the colour.
Qiskit (supported by Python 3.5 or later)
Get it by running the following command
pip install qiskit
Pygame
Get it by running the following command
pip install pygame
Run the Random_colours.py file on the IBM Quantum Experience. This should take about 5 minutes to give you a result, since we are generating 40,000 random numbers using the quantum circuit!
Copy and paste the output into the generated_cols list in the file QuantumLogo.py.
Run! You should find a file named Colour.jpeg saved in the directory where QuantumLogo.py is saved.
The current circuit is set to a probability of 25%, 37.5% and 37.5% for r,g,b respectively. That means that the generated image should approximate this colour, which has an rgb value of (64, 96, 96):
This is what we get:
It's not a great result, but I was still proud :)
Feel free to play around with the circuit to approximate the colour that you want by checking its rgb code and computing the relative measure of each r/g/b value and mapping it to an output state. You can also change the order of elements in clrs for convenience. (It's currently set to [r,g,b]).
You can increase the accuracy of the generated colour to what you want by increasing the 'resolution'. You can do this by increasing the number of random numbers generated by tweaking the code in a couple of places.
Have fun!