Con las bombas que tiran los fanfarrones, se hacen las gaditanas tirabuzones (Palma y corona, Carmen Linares)
This time I draw Franky again using an algorithm to solve the Travelling Salesman Problem as I did in my last post. On this occasion, instead of doing just one single line drawing, I overlap many of them (250 concretely), each of them sampling 400 points on the original image (in my previous post I sampled 8.000 points). Last difference is that I don’t convert the image to pure black and white with threshold
function: now I use the gray scale number of each pixel to weight the sample.
Once again, I use ggplot2
package, and its magical geom_path
, to generate the image. The pencil effect is obtained giving a very high transparency to the lines. This is the result:
I love when someone else experiment with my experiments as Mara Averick did:
💼 single-line aRt w/ @aschinchon:
👨🎨 "The Travelling Salesman Portrait" https://t.co/MZqXeCleuU #rstats
(you'll never guess which are mine!) pic.twitter.com/CxXzZh7YRR— Mara Averick (@dataandme) April 5, 2018
An image as a travelling salesman problem using #rstats. How did I do it? I didn't: I just cloned a #GitHub repo (https://t.co/u7GDUzsJRu) and put in my own picture. Really nice, follow his blog (https://t.co/rmOqmmsd6T) 👍 pic.twitter.com/zLYXokR6U7
— Erik-Jan van Kesteren (@ejvankesteren) April 12, 2018
You can do it as well with this one, since you will find the code here. Please, let me know your own creations if you do. You can find me on twitter or by email.
P.S.: Although it may seems otherwise, I’m not obsessed with Frankenstein 🙂