Clear Filters
Clear Filters

Color zooming or halftoning

3 views (last 30 days)
Mesho
Mesho on 23 Oct 2023
Commented: Mesho on 26 Oct 2023
Dear all,
I have this image from wikipedia that explains halftoning.
I want to experiment the same thing with matlab for other color combinations. So what will be the output color I will see from a far distance when combining different dots with different size and rotations?
Any suggestion will be appreciated.

Accepted Answer

Walter Roberson
Walter Roberson on 24 Oct 2023
Decided on a per-layer transparency, something like
transparency = 0.8;
Create each halftone in an array.
image() the lowest layer (the one to go underneath)
hold on
now loop over the other layers:
alphadata = transparency .* double(thislayer ~= 0);
image(thislayer, 'AlphaData', alphadata);
At the end of the loop, turn hold off
  3 Comments
Walter Roberson
Walter Roberson on 25 Oct 2023
The potential problem with zooming it out is that you are then subject to whatever graphics smoothing that MATLAB chooses to do. You can turn off the explicit smoothing, https://www.mathworks.com/help/matlab/creating_plots/using-graphics-smoothing.html but at some point it is going to have to interpolate data from pixels to fit into the physical screen.
You might well need 200 or more dots per inch https://www.atlasscreensupply.com/blog/choosing-the-right-mesh-and-halftone-dot.htm for your halftone. Ah, more values at https://en.wikipedia.org/wiki/Halftone#Resolution_of_halftone_screens -- you will probably need more than 300 dots per inch.
Typical desktop screen resolution does not get that fine. My relative high resolution iMac, 5120 x 2880, is about 218 dots per inch.
Some laptops do get up to those resolutions -- and some smartphones do get up there. For example the recent iPhones have had resolution of roughly 460 dots per inch... on a screen that is about 6.5 inches diagonal.
If the number of pixels in your halftone images exceeds the physical limits of the display you are using, then some layer is going to have to reduce information down. So to avoid that, do not bother with an image that is more than roughly 1000 x 1200 /
Mesho
Mesho on 26 Oct 2023
Didn't run the experiment yet but thank you for the comments.

Sign in to comment.

More Answers (0)

Categories

Find more on Image Processing Toolbox in Help Center and File Exchange

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!