mlib_ImageColorDitherInit

- initialization for image dithering

Synopsis

cc [ flag... ] file... -lmlib [ library... ]
#include <mlib.h>

mlib_status mlib_ImageColorDitherInit(void **colormap, const mlib_s32 *dimensions,
     mlib_type intype, mlib_type outtype, mlib_s32 channels, mlib_s32 entries,
     mlib_s32 offset, void **lut);

Description

The mlib_ImageColorDitherInit() function creates an internal data structure, colormap, which can be used by one of the following functions for image dithering:

mlib_ImageColorErrorDiffusion3x3
mlib_ImageColorErrorDiffusionMxN
mlib_ImageColorOrderedDither8x8
mlib_ImageColorOrderedDitherMxN

The lut might have either 1 or 3 channels. The type of the lut can be one of the following:

MLIB_BYTE in, MLIB_BYTE out (i.e., BYTE-to-BYTE)
MLIB_BIT in, MLIB_BYTE out (i.e., BIT-to-BYTE)

If dimensions == NULL, then no colorcube will be created. In this case, the user-provided lookup table, lut, will be used for dithering.

If dimensions != NULL, then a colorcube is created from scratch in a way shown in the following example.

To dither an RGB image of type MLIB_BYTE to a color-indexed image of type MLIB_BYTE, we can use the following parameters:

mlib_s32  dimensions[] = {2, 3, 4};
mlib_type intype = MLIB_BYTE;
mlib_type outtype = MLIB_BYTE;
mlib_s32  channels = 3;
mlib_s32  offset = 6;

These values would lead to the creation of a colorcube that would dither red values in the source image to one of 2 red levels, green values to one of 3 green levels, and blue values to one of 4 blue levels. You could picture this colorcube as a cube with dimensions of 2, 3, and 4. The index values assigned to the elements in that cube can be described by the following lookup table:

multipliers[0] = signof(dimensions[0])*1;
multipliers[i] = signof(dimensions[i])*
                 abs(multipliers[i-1]*dimension[i-1]);

A negative dimensions[i], so as to a negative multipliers[i], indicates that the values in a color ramp for channel i should appear in decreasing as opposed to increasing order.

For each channel i, the values of the levels are determined by the following formulas:

double delta = (dataMax - dataMin)/(abs(dimensions[i]) - 1);
int levels[j] = (int)(j*delta + 0.5);

where dataMax and dataMin are the maximum and minimum values, respectively, for data type intype.

Whenever a colorcube is created, if lut != NULL, the lookup table will be filled according to the colorcube and supplied parameters like offset. For the example shown above, the lookup table will start from line 6. In this case, it is the user's responsibility to allocate memory for the lookup table.

Parameters

The function takes the following arguments:

colormap

Internal data structure for image dithering.

dimensions

Dimensions of the colorcube in the colormap structure.

intype

Data type of the source image and the lookup table.

outtype

Data type of the destination indexed image.

channels

Number of channels of the lookup table and source image.

entries

Number of entries of the lookup table.

offset

Index offset of the lookup table.

lut

Lookup table.

Return Values

The function returns MLIB_SUCCESS if successful. Otherwise it returns MLIB_FAILURE.

Attributes

See attributes(5) for descriptions of the following attributes:

See Also

mlib_ImageColorDitherFree(3MLIB), mlib_ImageColorErrorDiffusion3x3(3MLIB), mlib_ImageColorErrorDiffusionMxN(3MLIB), mlib_ImageColorOrderedDither8x8(3MLIB), mlib_ImageColorOrderedDitherMxN(3MLIB), attributes(5)