diff options
author | Tavian Barnes <tavianator@gmail.com> | 2009-03-20 04:06:39 +0000 |
---|---|---|
committer | Tavian Barnes <tavianator@gmail.com> | 2009-03-20 04:06:39 +0000 |
commit | 047d3248896d375a8fbc80dbbf573b81a3e5a927 (patch) | |
tree | cdfdeebf0477ac62711683a5c267235f0a286ecd /libdimension/color.c | |
parent | 5ac6158f8ff999d4db18fb805c02b5c733e75ddb (diff) | |
download | dimension-047d3248896d375a8fbc80dbbf573b81a3e5a927.tar.xz |
Revamp color handling.
Diffstat (limited to 'libdimension/color.c')
-rw-r--r-- | libdimension/color.c | 259 |
1 files changed, 259 insertions, 0 deletions
diff --git a/libdimension/color.c b/libdimension/color.c new file mode 100644 index 0000000..bbde48d --- /dev/null +++ b/libdimension/color.c @@ -0,0 +1,259 @@ +/************************************************************************* + * Copyright (C) 2008 Tavian Barnes <tavianator@gmail.com> * + * * + * This file is part of Dimension. * + * * + * Dimension is free software; you can redistribute it and/or modify it * + * under the terms of the GNU Lesser General Public License as published * + * by the Free Software Foundation; either version 3 of the License, or * + * (at your option) any later version. * + * * + * Dimension is distributed in the hope that it will be useful, but * + * WITHOUT ANY WARRANTY; without even the implied warranty of * + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU * + * Lesser General Public License for more details. * + * * + * You should have received a copy of the GNU Lesser General Public * + * License along with this program. If not, see * + * <http://www.gnu.org/licenses/>. * + *************************************************************************/ + +#include "dimension.h" +#include <math.h> /* For pow() */ + +dmnsn_CIE_XYZ whitepoint = { 0.9505, 1, 1.089 }; + +dmnsn_color +dmnsn_color_from_XYZ(dmnsn_CIE_XYZ XYZ) +{ + dmnsn_color ret = { XYZ.X, XYZ.Y, XYZ.Z, 0.0, 0.0 }; + return ret; +} + +dmnsn_color +dmnsn_color_from_xyY(dmnsn_CIE_xyY xyY) +{ + dmnsn_color ret = { xyY.Y*xyY.x/xyY.y, + xyY.Y, + xyY.Y*(1.0 - xyY.x - xyY.Y)/xyY.y, + 0.0, 0.0 }; + return ret; +} + +static double Lab_finv(double t) { + if (t > 6.0/29.0) { + return t*t*t; + } else { + return 108.0*(t - 16.0/116.0)/841.0; + } +} + +dmnsn_color +dmnsn_color_from_Lab(dmnsn_CIE_Lab Lab, dmnsn_CIE_XYZ white) +{ + double fx, fy, fz; + dmnsn_color ret; + + fy = (Lab.L + 16.0)/116.0; + fx = fy + Lab.a/500.0; + fz = fy - Lab.b/200.0; + + ret.X = white.X*Lab_finv(fx); + ret.Y = white.Y*Lab_finv(fy); + ret.Z = white.Z*Lab_finv(fz); + + return ret; +} + +dmnsn_color +dmnsn_color_from_Luv(dmnsn_CIE_Luv Luv, dmnsn_CIE_XYZ white) +{ + double fy; + double uprime, unprime, vprime, vnprime; + dmnsn_color ret; + + fy = (Luv.L + 16.0)/116.0; + + unprime = 4.0*white.X/(white.X + 15.0*white.Y + 3.0*white.Z); + uprime = Luv.u/Luv.L/13.0 + unprime; + vnprime = 9.0*white.Y/(white.X + 15.0*white.Y + 3.0*white.Z); + vprime = Luv.v/Luv.L/13.0 + vnprime; + + ret.Y = white.Y*Lab_finv(fy); + ret.X = ret.Y*9.0*uprime/vprime/4.0; + ret.Z = ret.Y*(12.0 - 3*uprime - 20*vprime)/vprime/4.0; + + return ret; +} + +dmnsn_color +dmnsn_color_from_sRGB(dmnsn_sRGB sRGB) +{ + double Rlinear, Glinear, Blinear; /* Linear RGB values - no gamma */ + dmnsn_color ret; + + /* + * If C represents R, G, and B, then the Clinear values are now found as + * follows: + * + * { Csrgb/12.92, Csrgb <= 0.04045 + * Clinear = { 1/2.4 + * { ((Csrgb + 0.055)/1.055) , Csrgb > 0.04045 + */ + + if (sRGB.R <= 0.04045) { + Rlinear = sRGB.R/19.92; + } else { + Rlinear = pow((sRGB.R + 0.055)/1.055, 2.4); + } + + if (sRGB.G <= 0.04045) { + Glinear = sRGB.G/19.92; + } else { + Glinear = pow((sRGB.G + 0.055)/1.055, 2.4); + } + + if (sRGB.B <= 0.04045) { + Blinear = sRGB.B/19.92; + } else { + Blinear = pow((sRGB.B + 0.055)/1.055, 2.4); + } + + /* + * Now, the linear conversion. Expressed as matrix multiplication, it looks + * like this: + * + * [X] [0.4124 0.3576 0.1805] [Rlinear] + * [Y] = [0.2126 0.7152 0.0722]*[Glinear] + * [X] [0.0193 0.1192 0.9505] [Blinear] + */ + + ret.X = 0.4124*Rlinear + 0.3576*Glinear + 0.1805*Blinear; + ret.Y = 0.2126*Rlinear + 0.7152*Glinear + 0.0722*Blinear; + ret.Z = 0.0193*Rlinear + 0.1192*Glinear + 0.9505*Blinear; + ret.filter = 0.0; + ret.trans = 0.0; + + return ret; +} + +dmnsn_CIE_XYZ +dmnsn_XYZ_from_color(dmnsn_color color) +{ + dmnsn_CIE_XYZ ret = { color.X, color.Y, color.Z }; + return ret; +} + +dmnsn_CIE_xyY +dmnsn_xyY_from_color(dmnsn_color color) +{ + dmnsn_CIE_xyY ret = { color.X/(color.X + color.Y + color.Z), + color.Y/(color.X + color.Y + color.Z), + color.Y }; + return ret; +} + +static double Lab_f(double t) { + if (t > 216.0/24389.0) { + return pow(t, 1.0/3.0); + } else { + return 841.0*t/108.0 + 4.0/29.0; + } +} + +dmnsn_CIE_Lab +dmnsn_Lab_from_color(dmnsn_color color, dmnsn_CIE_XYZ white) +{ + dmnsn_CIE_Lab ret; + + ret.L = 116.0*Lab_f(color.Y/white.Y) - 16.0; + ret.a = 500.0*(Lab_f(color.X/white.X) - Lab_f(color.Y/white.Y)); + ret.b = 200.0*(Lab_f(color.Y/white.Y) - Lab_f(color.Z/white.Z)); + + return ret; +} + +dmnsn_CIE_Luv +dmnsn_Luv_from_color(dmnsn_color color, dmnsn_CIE_XYZ white) +{ + double uprime, unprime, vprime, vnprime; + dmnsn_CIE_Luv ret; + + uprime = 4.0*color.X/(color.X + 15.0*color.Y + 3.0*color.Z); + unprime = 4.0*white.X/(white.X + 15.0*white.Y + 3.0*white.Z); + vprime = 9.0*color.Y/(color.X + 15.0*color.Y + 3.0*color.Z); + vnprime = 9.0*white.Y/(white.X + 15.0*white.Y + 3.0*white.Z); + + ret.L = 116.0*Lab_f(color.Y/white.Y) - 16.0; + ret.u = 13.0*ret.L*(uprime - unprime); + ret.v = 13.0*ret.L*(vprime - vnprime); + + return ret; +} + +dmnsn_sRGB +dmnsn_sRGB_from_color(dmnsn_color color) +{ + double Rlinear, Glinear, Blinear; /* Linear RGB values - no gamma */ + dmnsn_sRGB ret; + + /* + * First, the linear conversion. Expressed as matrix multiplication, it looks + * like this: + * + * [Rlinear] [ 3.2410 -1.5374 -0.4986] [X] + * [Glinear] = [-0.9692 1.8760 0.0416]*[Y] + * [Blinear] [ 0.0556 -0.2040 1.0570] [Z] + */ + Rlinear = 3.2410*color.X - 1.5374*color.Y - 0.4986*color.Z; + Glinear = -0.9692*color.X + 1.8760*color.Y + 0.0416*color.Z; + Blinear = 0.0556*color.X - 0.2040*color.Y + 1.0570*color.Z; + + /* + * If C represents R, G, and B, then the sRGB values are now found as follows: + * + * { 12.92*Clinear, Clinear <= 0.0031308 + * Csrgb = { 1/2.4 + * { (1.055)*Clinear - 0.055, Clinear > 0.0031308 + */ + + if (Rlinear <= 0.0031308) { + ret.R = 12.92*Rlinear; + } else { + ret.R = 1.055*pow(Rlinear, 1.0/2.4) - 0.055; + } + + if (Glinear <= 0.0031308) { + ret.G = 12.92*Glinear; + } else { + ret.G = 1.055*pow(Glinear, 1.0/2.4) - 0.055; + } + + if (Blinear <= 0.0031308) { + ret.B = 12.92*Blinear; + } else { + ret.B = 1.055*pow(Blinear, 1.0/2.4) - 0.055; + } + + return ret; +} + +dmnsn_color +dmnsn_color_add(dmnsn_color color1, dmnsn_color color2) +{ + dmnsn_CIE_Lab Lab, Lab1, Lab2; + dmnsn_color ret; + + Lab1 = dmnsn_Lab_from_color(color1, whitepoint); + Lab2 = dmnsn_Lab_from_color(color2, whitepoint); + + Lab.L = Lab1.L + Lab2.L; + Lab.a = (Lab1.L*Lab1.a + Lab2.L*Lab2.a)/Lab.L; + Lab.b = (Lab1.L*Lab1.b + Lab2.L*Lab2.b)/Lab.L; + + ret = dmnsn_color_from_Lab(Lab, whitepoint); + ret.filter = (Lab1.L*color1.filter + Lab2.L*color2.filter)/Lab.L; + ret.trans = (Lab1.L*color1.trans + Lab2.L*color2.trans)/Lab.L; + + return ret; +} |