I'm available for new projects from August 2026, get in touch!

Simon Whittaker House of Deadleg Illustration & Design

Random sprout generation

Published December 8, 2014

For my Nana Loves Sprouts card I wanted to have the Nana presenting a huge bowl of sprouts. In order to make that bowl of sprouts there are a couple of ways I could go about it. Draw out hundreds of individual sprouts by hand. Draw out a couple of sprouts and then duplicate, rotate, resize, them etc to add a bit of variety.

In the end I decided to go with option 2, but with a twist. Instead of duplicating, rotating, etc myself, I'd make a generative art program to do it for me. I've posted previously about the Joshua Davis Skillshare class I took about just this kind of thing. I'm not sure generating random sprouts is quite what he had in mind when he set up the class mind.


  1. At this point in the image, everything is done except the bowl of sprouts.
  2. I created a placeholder shape to define the area I want to fill. I copied this and pasted it into Photoshop and saved it as a PNG with a transparent background. In Processing I'm going to use this to restrict where the program places the artwork – everything that's grey will be covered by sprouts.
  3. Back in Illustrator, I created two different sprouts to use to fill the bowl. I saved these out as separate, ungrouped SVG's, 50px by 50px.
  4. The two sprout SVG's and the PNG placeholder image are linked into the Processing file and the canvas is set to the same size as the PNG. I made a range of colours by creating a blend between two different greens, expanded it and extracting the colour values. The program will pick a random sprout SVG, use random colours from the selected range to colour each bit of artwork (as the SVGs are ungrouped each leaf will be coloured separately), applies a random rotation and scale and then places the artwork on the canvas. It took a couple of attempts to get the best coverage, but in the end the program was set to do this process 120 times. Every time its run the result is different. I used two different run-through's in the final artwork.
  5. After the program is run the output is saved as a PDF that looks something like this.
  6. The artwork is cut off at the edges, but opening it in Illustrator reveals that the whole artwork is there.
  7. A bit of editing and then the artwork is grouped, copied and then dropped into the main artwork and tweaked to fit.
  • 1. At this point in the image, everything is done except the bowl of sprouts.
  • 2. I created a placeholder shape to define the area I want to fill. I copied this and pasted it into Photoshop and saved it as a PNG with a transparent background. In Processing I'm going to use this to restrict where the program places the artwork – everything that's grey will be covered by sprouts.
  • 2. I created a placeholder shape to define the area I want to fill. I copied this and pasted it into Photoshop and saved it as a PNG with a transparent background. In Processing I'm going to use this to restrict where the program places the artwork – everything that's grey will be covered by sprouts.
  • 3. Back in Illustrator, I created two different sprouts to use to fill the bowl. I saved these out as separate, ungrouped SVG's, 50px by 50px
  • 3. Back in Illustrator, I created two different sprouts to use to fill the bowl. I saved these out as separate, ungrouped SVG's, 50px by 50px
  • 4. The two sprout SVG's and the PNG placeholder image are linked into the Processing file and the canvas is set to the same size as the PNG. I made a range of colours by creating a blend between two different greens, expanded it and extracting the colour values. The program will pick a random sprout SVG, use random colours from the selected range to colour each bit of artwork (as the SVGs are ungrouped each leaf will be coloured separately), applies a random rotation and scale and then places the artwork on the canvas. It took a couple of attempts to get the best coverage, but in the end the program was set to do this process 120 times. Every time its run the result is different. I used two different run-through's in the final artwork.
  • 5. After the program is run the output is saved as a PDF that looks something like this.
  • 6. The artwork is cut off at the edges, but opening it in Illustrator reveals that the whole artwork is there.
  • 7. A bit of editing and then the artwork is grouped, copied and then dropped into the main artwork and tweaked to fit.

What did you think?

If you enjoyed seeing a bit of a peek at my process – or even if you didn't – it'd be great if you'd let me know. If you could take a couple of minutes to let me know your thoughts it'd really be appreciated.

You can drop me a quick message on Twitter, Facebook or Instagram if that's your thing, or even an old fashioned(?) !

Thanks for stopping by:)


“Beyond my expectations”

Gordon Tennent

Sneakytrick

All content © Simon Whittaker/House of Deadleg 2026 unless otherwise specified.

Content/images may not be used in any automated data mining, web scraping, AI training or other processes without express permission.