Package net.sourceforge.plantuml.png.quantx

Class QuantizerWuOKLab

java.lang.Object

net.sourceforge.plantuml.png.quantx.QuantizerWuOKLab

public final class QuantizerWuOKLab
extends Object

Wu quantization over OKLab (L, a, b) instead of RGB. - 3D histogram on L, a, b (32 bins per axis -> INDEX_BITS = 5, INDEX_COUNT = 33 for integrals) - Cumulative moments/sums along each axis + squared sum (L^2 + a^2 + b^2) - Recursive splitting (maximize variance) same as the original algorithm - Average cube colors converted back to 8-bit sRGB Notes: - L in [0,1], a,b ~ [-0.5, 0.5] for sRGB; we clamp a,b to these limits - “Moments” L/a/b are stored as double; weights remain int

Constructor Summary

Constructors

Constructor	Description
QuantizerWuOKLab()

Method Summary

Modifier and Type	Method	Description
Map<Integer,Integer>	quantize(int[] pixels, int colorCount)

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Details

QuantizerWuOKLab

public QuantizerWuOKLab()

Method Details

quantize

public Map<Integer,Integer> quantize(int[] pixels,
 int colorCount)