To LUGNET HomepageTo LUGNET News HomepageTo LUGNET Guide Homepage
 Help on Searching
 
Post new message to lugnet.build.mosaicOpen lugnet.build.mosaic in your NNTP NewsreaderTo LUGNET News Traffic PageSign In (Members)
 Building / Mosaics / 365
364  |  366
Subject: 
Re: Mosaic programs
Newsgroups: 
lugnet.build.mosaic
Date: 
Tue, 15 Feb 2005 18:22:21 GMT
Viewed: 
5429 times
  
In lugnet.build.mosaic, Maarten Steurbaut wrote:
   Now we have something to compare:



On the left the original image.
In the middle my result with black, white, yellow, red and blue.
On the right Corey’s result with the same colours.

I don’t make artistic mosaics, they’re purely mathematical. I use the Floyd-Steinberg dithering algorythm. Bob uses it too (plus a few variations of this algorythm).
So Corey: do you use an existing algorythm (like iterative algorythms, that I can’t program unfortunately), or have you invented something completely new?
The first thing I noticed about your image: the area bottom-right is entirely white, while this area on the original image isn’t. With my FS-algorythm I expected pixels of a different colour in this area (I didn’t remove them manually to make comparison possible). What did you do to get this area white? Do you use some kind of “less dithering”-method?
In my mosaics there are many 1x1 elements, much more than in yours. Is this intended or just a coincidence? I’m asking this because I also make ministeck mosaics where 1x1 elements are very rare. I wrote an extra module to avoid as many 1x1 elements as possible, but I’m not completely happy about it. All suggestions are welcome.
Maybe you can lift a corner of your veil?

I’d say it looks like the image on the right was contrast enhanced before the dithering was applied. Stand back and squint at the images to see the change in contrast. The light colors are all clearly whiter. You can also see some color banding beneath the dither on the top right of the forehead, which is what you’d expect if some sort of histogram filter was applied to stretch the range of colors in the image before dithering. And although I don’t see evidence of it here, you might want to apply an edge enhancement filter before dithering to make the resulting mosaic look crisper.

Have fun,

Don



Message is in Reply To:
  Mosaic programs
 
Now we have something to compare: (URL) On the left the original image. In the middle my result with black, white, yellow, red and blue. On the right Corey's result with the same colours. I don't make artistic mosaics, they're purely mathematical. I (...) (20 years ago, 15-Feb-05, to lugnet.build.mosaic, FTX)

2 Messages in This Thread:

Entire Thread on One Page:
Nested:  All | Brief | Compact | Dots
Linear:  All | Brief | Compact
    

Custom Search

©2005 LUGNET. All rights reserved. - hosted by steinbruch.info GbR