Using K-means clustering it is possible to reduce the amount of colors in an image. What are the purposes of using this method for extracting colors from an image?
$\begingroup$
$\endgroup$
2
-
$\begingroup$ one purpose is palette generation. please review How to Ask and present your research. $\endgroup$– Christoph RackwitzJan 8 at 21:47
-
$\begingroup$ Note that there are other methods much more suitable than k-means for finding a reduced palette. $\endgroup$– Cris LuengoJul 14 at 17:00
Add a comment
|
1 Answer
$\begingroup$
$\endgroup$
There are some use cases:
- Palette Generation
The centroids becomes a color in the palette. - Segmentation
If you add the location of the pixels as a feature you can use the K-Means as segmentation tool as done in Super Pixels (See Rectangle Segments of Image (Rectangle Super Pixels) per Pixel and Locate Non Homogenous Areas in an Image).