%
% LanczosSharp is a slightly sharpened (blur=0.9830391168464) form of
% the Lanczos filter. It was designed specifically for cylindrical
-% (radial) EWA (Elliptical Weighted Average) distortion (using a
-% Jinc windowed Jinc), but can also be used as a slightly sharper
-% orthogonal Lanczos (Sinc-Sinc) filter.
+% (radial) EWA (Elliptical Weighted Average) distortion (using a Jinc
+% windowed Jinc); it can also be used as a slightly sharper
+% orthogonal Lanczos (Sinc-Sinc) filter. With the chosen blur value,
+% the corresponding EWA filter comes as close as possible to
+% satisfying the following condition:
+%
+% 'No-Op' Vertical and Horizontal Line Preservation Condition:
+% Images with only vertical or horizontal features are preserved
+% when performing 'no-op" with EWA distortion.
%
% Two forms of the 'Sinc' function are available: Sinc and SincFast.
% Sinc is computed using the traditional sin(pi*x)/(pi*x); it is
% and rational (high Q) approximations, and will be used by default in
% most cases.
%
-% The Lanczos2D and Lanczos2DSharp filters specific for EWA distorts.
-% They are both 2 lobe Lanczos-like filter but always using a
-% Jinc-Jinc filter function regardless of if used as a orthoginal
-% filter or not. The 'sharp' version has been sharpened using a
-% blur=0.958033808.
-%
-% Robidoux is a unique Keys cubic spline, developed from the previous
-% Lanczos2DSharp filter. It is designed to satisfy the following
-% condition:
-%
-% Robidoux exactly preserves images with only vertical or
-% horizontal features when performing 'no-op" with EWA distortion.
-%
-% That is, Robidoux is the BC-Spline with B=(228 - 108 sqrt(2))/199
-% and C=(108 sqrt(2) - 29)/398. Robidoux turns out to be close to
-% both plain Mitchell and "sharpened" Lanczos2D. For example, it's
-% first crossing is (36 sqrt(2) + 123)/(72 sqrt(2) + 47) which is
-% almost identical to the first crossing of the other two.
+% The Lanczos2D and Lanczos2DSharp filters are for EWA distorts.
+% They are both 2-lobe Lanczos-like filters but always use a
+% Jinc-Jinc filter function regardless of whether or not they are
+% used as orthogonal filters. The 'sharp' version uses
+% blur=0.958033808, chosen because the resulting EWA filter comes as
+% close as possible to satisfing the "'No-Op' Vertical and Horizontal
+% Line Preservation Condition".
+%
+% Robidoux is another filter tuned for EWA. It is the Keys cubic
+% spline defined by B=(228 - 108 sqrt(2))/199. Robidoux satisfies the
+% "'No-Op' Vertical and Horizontal Line Preservation Condition"
+% exactly. It turns out to be close to both plain Mitchell and
+% "sharpened" Lanczos2D; for example, its first crossing is (36
+% sqrt(2) + 123)/(72 sqrt(2) + 47) which is almost identical to the
+% first crossing of the other two.
%
% 'EXPERT' OPTIONS:
%
window;
} const mapping[SentinelFilter] =
{
- { UndefinedFilter, BoxFilter }, /* Undefined (default to Box) */
- { PointFilter, BoxFilter }, /* SPECIAL: Nearest neighbour */
- { BoxFilter, BoxFilter }, /* Box averaging filter */
- { TriangleFilter, BoxFilter }, /* Linear interpolation filter */
- { HermiteFilter, BoxFilter }, /* Hermite interpolation filter */
- { SincFastFilter, HanningFilter }, /* Hanning -- cosine-sinc */
- { SincFastFilter, HammingFilter }, /* Hamming -- '' variation */
+ { UndefinedFilter, BoxFilter }, /* Undefined (default to Box) */
+ { PointFilter, BoxFilter }, /* SPECIAL: Nearest neighbour */
+ { BoxFilter, BoxFilter }, /* Box averaging filter */
+ { TriangleFilter, BoxFilter }, /* Linear interpolation filter */
+ { HermiteFilter, BoxFilter }, /* Hermite interpolation filter */
+ { SincFastFilter, HanningFilter }, /* Hanning -- cosine-sinc */
+ { SincFastFilter, HammingFilter }, /* Hamming -- '' variation */
{ SincFastFilter, BlackmanFilter }, /* Blackman -- 2*cosine-sinc */
- { GaussianFilter, BoxFilter }, /* Gaussian blur filter */
- { QuadraticFilter, BoxFilter }, /* Quadratic Gaussian approx */
- { CubicFilter, BoxFilter }, /* Cubic B-Spline */
- { CatromFilter, BoxFilter }, /* Cubic-Keys interpolator */
- { MitchellFilter, BoxFilter }, /* 'Ideal' Cubic-Keys filter */
+ { GaussianFilter, BoxFilter }, /* Gaussian blur filter */
+ { QuadraticFilter, BoxFilter }, /* Quadratic Gaussian approx */
+ { CubicFilter, BoxFilter }, /* Cubic B-Spline */
+ { CatromFilter, BoxFilter }, /* Cubic-Keys interpolator */
+ { MitchellFilter, BoxFilter }, /* 'Ideal' Cubic-Keys filter */
{ LanczosFilter, SincFastFilter }, /* SPECIAL: 3-lobed sinc-sinc */
- { JincFilter, BoxFilter }, /* Raw 3-lobed Jinc function */
- { SincFilter, BoxFilter }, /* Raw 4-lobed Sinc function */
- { SincFastFilter, KaiserFilter }, /* Kaiser -- square root-sinc */
- { SincFastFilter, WelshFilter }, /* Welsh -- parabolic-sinc */
- { SincFastFilter, CubicFilter }, /* Parzen -- cubic-sinc */
- { LagrangeFilter, BoxFilter }, /* Lagrange self-windowing */
- { SincFastFilter, BohmanFilter }, /* Bohman -- 2*cosine-sinc */
+ { JincFilter, BoxFilter }, /* Raw 3-lobed Jinc function */
+ { SincFilter, BoxFilter }, /* Raw 4-lobed Sinc function */
+ { SincFastFilter, KaiserFilter }, /* Kaiser -- square root-sinc */
+ { SincFastFilter, WelshFilter }, /* Welsh -- parabolic-sinc */
+ { SincFastFilter, CubicFilter }, /* Parzen -- cubic-sinc */
+ { LagrangeFilter, BoxFilter }, /* Lagrange self-windowing */
+ { SincFastFilter, BohmanFilter }, /* Bohman -- 2*cosine-sinc */
{ SincFastFilter, TriangleFilter }, /* Bartlett -- triangle-sinc */
- { SincFastFilter, BoxFilter }, /* Raw fast sinc ("Pade"-type) */
+ { SincFastFilter, BoxFilter }, /* Raw fast sinc ("Pade"-type) */
{ LanczosSharpFilter, SincFastFilter }, /* SPECIAL: Sharpened Lanczos */
- { Lanczos2DFilter, JincFilter }, /* SPECIAL: 2-lobed jinc-jinc */
- { Lanczos2DSharpFilter, JincFilter }, /* SPECIAL: ditto sharpened */
- { RobidouxFilter, BoxFilter }, /* SPECIAL: Keys tuned for EWA */
+ { Lanczos2DFilter, JincFilter }, /* SPECIAL: 2-lobed jinc-jinc */
+ { Lanczos2DSharpFilter, JincFilter }, /* SPECIAL: ditto sharpened */
+ { RobidouxFilter, BoxFilter }, /* SPECIAL: Keys tuned for EWA */
};
/*
Table mapping the filter/window from the above table to an actual
filter_type=JincFilter;
break;
case LanczosSharpFilter:
- /* A slightly sharper Lanczos */
+ /* Sharpen Lanczos so as to optimize for minimal blurring of
+ * orthogonal lines in EWA cylindrical usage. Value
+ * determined by Nicolas Robidoux.
+ */
resize_filter->blur *= 0.9830391168464;
/* fall-thru to promote it to Jinc-Jinc */
case LanczosFilter:
resize_filter->blur *= 0.958033808;
break;
case GaussianFilter:
- sigma = (MagickRealType) (MagickSQ2/2.0); /* Cylindrical Gaussian sigma is sqrt(2)/2 */
+ /* Cylindrical Gaussian sigma is sqrt(2)/2. */
+ sigma = (MagickRealType) (MagickSQ2/2.0);
break;
default:
break;
switch (filter_type)
{
case LanczosSharpFilter:
- resize_filter->blur *= 0.9830391168464; /* Orthogonal sharpen too */
+ resize_filter->blur *= 0.9830391168464; /* Orthogonal sharpen too. */
break;
case Lanczos2DSharpFilter:
- resize_filter->blur *= 0.958033808; /* Orthogonal sharpen too */
+ resize_filter->blur *= 0.958033808; /* Orthogonal sharpen too. */
break;
default:
break;
if (artifact != (const char *) NULL)
{
C=StringToDouble(artifact);
- B=1.0-2.0*C; /* Calculate B to get a Keys cubic filter. */
+ B=1.0-2.0*C; /* Calculate B to get a Keys cubic filter. */
}
}
/* Convert B,C values into Cubic Coefficents. See CubicBC(). */
projects / imagemagick / commitdiff