#ALGORITHM GENETIC - REPLICED IMAGE

This project is based on Alteredqualia's project.

The goal is to get an image represented as a collection of overlapping circles of various colors and transparencies.

FEATURES

START:

1. We start from random number of circle (1 to n) on a superfice.
2. Each gene has

MUTATE:

1. In each step we randomly modify one or more parameters of circle (like color, position or radius), addfy or removefy a circle;
2. The number of genes can be change is a random number (1 to 20% of genes length).

CANVAS:

1. The target image is showed in main canvas;
2. The best individual is showed in secound canvas;
3. The project has a hidden canvas for calculate fitness;
4. User can print image (print button);
5. The application save images when the fitness of best image has reduced 2.000.000;
6. Imagens are save in data/NAME_OF_PROJECT folder.

FITNESS:

1. The function fitness is calculate with sum of difference of all points;
2. This difference is calculated with this equation: |r1 r2| + |g1 + g2| + |b1 + b2|.

CUSTOM

1. Users can choose background color (white, black or hex color);
2. Users can change maximus radius of circle;
3. Users can select specific image;
4. User can change number maximum initial of genes;
5. User can choose color mode of mutation (color or grayscale).

RESULTS

How does it look after some time?

Does it work on all images?

CONTACT

If you have any questions, suggestions, comments, you can send me a mail (luisaraujo.ifba@gmail.com). Feel free to improve on this implementation.

Source is released under MIT License.