Weaver

The idea is to take an image and come up with data to weave it using a single thread on a circular rim.

Basically this: A picture is worth a thousand words.

Actual result

Output	Input

No. of pins: 254

No. of lines: 4000 (Approx. 1.75 km thread)

Credit: @larryisgood

Screenshots

Multiple colors

Grayscale

Basic algorithm

This piece of software requires 3 inputs from the user:

A picture to be drawn
Total points/nails on a circular rim, P
Total lines to be drawn, L

All three inputs are crucial on how the final result is going to look. The program does the following:

Converts the color picture to grayscale.
The picture is then cropped to a square.
Places a virtual circular rim with P equidistant points on it.
Takes the first point (p = 0) as the starting point.
Finds the next point p' to draw a line from point p by finding the highest intensity line.
- The intensity of a line is calculated by adding up the all the pixel values of a line from p to p'.
A new line is drawn from p to p' on the circular rim.
The original image is modified such that the pixels under the line from p to p' are lightened* so that the same line is not drawn next time.
p' is set as the new starting point p and then the steps from 5 is repeated until L lines are drawn.

* The amount by which a line should be lightened can also be configured.

How to run?

You'll need to download Processing. It's available for Windows / Mac / Linux.
Open up Weaver.pde using Processing.
Hit Run or press CTRL+R to execute the application.
Click anywhere on the application to start processing.
Click again to pause.
After the process is complete, a text file will be generated. This file contains the positions of each point where the thread should be knitted next.

Note: Those of you who are downloading the repository as .zip file, you must rename the folder Weaver-master to Weaver before opening on Processing.

Configuration

The setup() function in Weaver.pde chooses one of the example configurations. Check the example.pde to modify the configuration of each example.

Credits

Kudos to i-make-robots for originally writting this algorithm.

I changed it a little bit to clean up the code, improve GUI and add some extra features.

alyyousuf7/Weaver