Overview

This program takes in an image and transforms it into an abstract neon image according to the image’s contours. The resulting image will be different for 2 consecuitive runs with the program since many random numbers are used in generating the final output.

Input

After running the program with the provided example image, you will get a result that is similar to the one below (but not exactly the same).

Output

The source code for this program is also hosted on GitHub at the following link:

https://github.com/jjones646/mkneon

Workflow

Convert to grayscale
Randomly threshold image multiple times
Clean up thresholded image with erode & dilate
Use OpenCV’s findContours() API
Iterate over each set of contours & apply a random stroke width and a random color

Credits

Extracting the contour lines out of an image is heavily based on snippets of code from Adrian Rosebrock’s blog post that can be found here:

http://www.pyimagesearch.com/2016/04/11/finding-extreme-points-in-contours-with-opencv