Image processing method and apparatus

Claim 2

An image processing apparatus for quantizing multilevel color image data containing at least two kinds of color components, comprising: error addition means for adding a quantization error value distributed from a neighboring pixel for each color component to each of a plurality of color components contained in a target pixel; a threshold table which stores a quantization threshold for each of the color components in accordance with a combination of color component values of pixels; modulation amount determination means for determining a threshold modulation amount of each color in accordance with a combination of color component values of pixels including the target pixel; and quantization means for acquiring a threshold for each color component from said threshold table in accordance with a combination of color components of the target pixel, determining a threshold modulated by adding the modulation amount to the threshold for each color, and quantizing the target pixel in accordance with a relationship in magnitude between the modulated threshold and a value of each color component to which an error value is added by said error addition means.

Claim 3

The apparatus according to claim 1 , wherein said modulation amount determination means determines a threshold modulation amount for each color in accordance with a combination of color component values of the target pixel.

Claim 4

The apparatus according to claim 1 , wherein said modulation amount determination means determines a threshold modulation amount for each color in accordance with a combination of average values, maximum values, or minimum values of color components of the target pixel and a neighboring pixel thereof.

Claim 5

An image processing apparatus for quantizing multilevel color image data containing at least two kinds of color components, comprising: error addition means for adding a quantization error value distributed from a neighboring pixel for each color component to each of a plurality of color components contained in a target pixel; quantization means for quantizing each color component of the target pixel; and a diffusion coefficient table which stores a diffusion coefficient for diffusing a quantization error produced by said quantization means, in accordance with a combination of color component values of a pixel, wherein said error addition means adds an error value to the target pixel in accordance with a combination of color components of the target pixel and a diffusion coefficient selected from said diffusion coefficient table.

Claim 6

An image processing apparatus for quantizing multilevel color image data containing at least two kinds of color components, comprising: error addition means for adding a quantization error value distributed from a neighboring pixel for each color component to each of a plurality of color components contained in a target pixel; quantization means for quantizing each color component of the target pixel; and a diffusion coefficient table which stores a diffusion coefficient for diffusing a quantization error produced by said quantization means, in accordance with a combination of color component values of a pixel, wherein said error addition means adds an error value to the target pixel in accordance with a combination of color components of the target pixel and a diffusion coefficient selected from said diffusion coefficient table.

Claim 7

The apparatus according to claim 4 , wherein said error addition means selects a diffusion coefficient from said diffusion coefficient table in accordance with a combination of color component values of the target pixel.

Claim 8

The apparatus according to claim 4 , wherein said error addition means selects a diffusion coefficient from said diffusion coefficient table in accordance with a combination of average values, maximum values, or minimum values of color components of the target pixel and a neighboring pixel thereof.

Claim 9

The apparatus according to claim 4 , wherein said error addition means selects a diffusion coefficient from said diffusion coefficient table in accordance with a product of color component values of the target pixel and a neighboring pixel thereof.

Claim 10

An image processing apparatus for quantizing multilevel color image data containing at least two kinds of color components, comprising: error addition means for adding a quantization error value distributed from a neighboring pixel for each color component to each of a plurality of color components contained in a target pixel; a threshold table which stores a quantization threshold for each of the color components in accordance with a combination of color component values of pixels; modulation amount determination means for determining a threshold modulation amount of each color in accordance with a combination of color component values of pixels including the target pixel; quantization means for acquiring a threshold for each color component from said threshold table in accordance with a combination of color components of the target pixel, determining a threshold modulated by adding the modulation amount to the threshold for each color, and quantizing the target pixel in accordance with a relationship in magnitude between the modulated threshold and a value of each color component to which an error value is added by said error addition means; and a diffusion coefficient table which stores a diffusion coefficient for diffusing a quantization error produced by said quantization means, in accordance with a combination of color component values of a pixel, wherein said error addition means adds an error value to the target pixel in accordance with a combination of color components of the target pixel and a diffusion coefficient selected from said diffusion coefficient table.

Claim 11

An image processing apparatus for quantizing multilevel color image data containing at least two kinds of color components, comprising: error addition means for adding a quantization error value distributed from a neighboring pixel for each color component to each of a plurality of color components contained in a target pixel; a threshold table which stores a quantization threshold for each of the color components in accordance with a combination of color component values of pixels; modulation amount determination means for determining a threshold modulation amount of each color in accordance with a combination of color component values of pixels including the target pixel; quantization means for acquiring a threshold for each color component from said threshold table in accordance with a combination of color components of the target pixel, determining a threshold modulated by adding the modulation amount to the threshold for each color, and quantizing the target pixel in accordance with a relationship in magnitude between the modulated threshold and a value of each color component to which an error value is added by said error addition means; and a diffusion coefficient table which stores a diffusion coefficient for diffusing a quantization error produced by said quantization means, in accordance with a combination of color component values of a pixel, wherein said error addition means adds an error value to the target pixel in accordance with a combination of color components of the target pixel and a diffusion coefficient selected from said diffusion coefficient table.

Claim 12

An image processing apparatus for quantizing multilevel color image data containing at least three kinds of color components, comprising: error addition means for adding a quantization error value distributed from a neighboring pixel for each color component to each of a plurality of color components contained in a target pixel; a threshold table which stores a quantization threshold for each of the color components in accordance with a combination of color component values of pixels; modulation amount determination means for determining a threshold modulation amount of each color in accordance with a combination of color component values of pixels including the target pixel; and quantization means for acquiring a threshold for each color component from said threshold table in accordance with a combination of color components of the target pixel, determining a threshold modulated by adding the modulation amount to the threshold for each color, and quantizing the target pixel in accordance with a relationship in magnitude between the modulated threshold and a value of each color component to which an error value is added by said error addition means, wherein a combination of two kinds of color components is quantized by said image processing apparatus and a remaining color component is quantized with a value of a color component other than a target color component being regarded as 0.

Claim 13

An image processing apparatus for quantizing multilevel color image data containing at least three kinds of color components, comprising: error addition means for adding a quantization error value distributed from a neighboring pixel for each color component to each of a plurality of color components contained in a target pixel; a threshold table which stores a quantization threshold for each of the color components in accordance with a combination of color component values of pixels; modulation amount determination means for determining a threshold modulation amount of each color in accordance with a combination of color component values of pixels including the target pixel; and quantization means for acquiring a threshold for each color component from said threshold table in accordance with a combination of color components of the target pixel, determining a threshold modulated by adding the modulation amount to the threshold for each color, and quantizing the target pixel in accordance with a relationship in magnitude between the modulated threshold and a value of each color component to which an error value is added by said error addition means, wherein a combination of two kinds of color components is quantized by said image processing apparatus and a remaining color component is quantized with a value of a color component other than a target color component being regarded as 0.

Claim 14

An image processing method of quantizing multilevel color image data containing at least two kinds of color components, comprising: an error addition step of adding a quantization error value distributed from a neighboring pixel for each color component to each of a plurality of color components contained in a target pixel; a modulation amount determination step of determining a threshold modulation amount of each color in accordance with a combination of color component values of pixels including the target pixel; and a quantization step of acquiring a threshold for each color component, in accordance with a combination of color components of the target pixel, from a threshold table which stores a quantization threshold for each of the color components in accordance with a combination of color component values of pixels, determining a threshold modulated by adding the modulation amount to the threshold for each color, and quantizing the target pixel in accordance with a relationship in magnitude between the modulated threshold and a value of each color component to which an error value is added in the error addition step.

Claim 15

An image processing method of quantizing multilevel color image data containing at least two kinds of color components, comprising: an error addition step of adding a quantization error value distributed from a neighboring pixel for each color component to each of a plurality of color components contained in a target pixel; a modulation amount determination step of determining a threshold modulation amount of each color in accordance with a combination of color component values of pixels including the target pixel; and a quantization step of acquiring a threshold for each color component, in accordance with a combination of color components of the target pixel, from a threshold table which stores a quantization threshold for each of the color components in accordance with a combination of color component values of pixels, determining a threshold modulated by adding the modulation amount to the threshold for each color, and quantizing the target pixel in accordance with a relationship in magnitude between the modulated threshold and a value of each color component to which an error value is added in the error addition step.

Claim 16

The method according to claim 10 , wherein in the modulation amount determination step, a threshold modulation amount for each color is determined in accordance with a combination of color component values of the target pixel.

Claim 17

The method according to claim 10 , wherein in the modulation amount determination step, a threshold modulation amount for each color is determined in accordance with a combination of average values, maximum values, or minimum values of color components of the target pixel and a neighboring pixel thereof.

Claim 18

An image processing method of quantizing multilevel color image data containing at least two kinds of color components, comprising: an error addition step of adding a quantization error value distributed from a neighboring pixel for each color component to each of a plurality of color components contained in a target pixel; and a quantization step of quantizing each color component of the target pixel wherein in the error addition step, an error value is added to the target pixel in accordance with a combination of color components of the target pixel and a diffusion coefficient selected from a diffusion coefficient table which stores a diffusion coefficient for diffusing a quantization error produced in the quantization step, in accordance with a combination of color component values of a pixel.

Claim 19

An image processing method of quantizing multilevel color image data containing at least two kinds of color components, comprising: an error addition step of adding a quantization error value distributed from a neighboring pixel for each color component to each of a plurality of color components contained in a target pixel; and a quantization step of quantizing each color component of the target pixel wherein in the error addition step, an error value is added to the target pixel in accordance with a combination of color components of the target pixel and a diffusion coefficient selected from a diffusion coefficient table which stores a diffusion coefficient for diffusing a quantization error produced in the quantization step, in accordance with a combination of color component values of a pixel.

Claim 20

The method according to claim 13 , wherein in the error addition step, a diffusion coefficient is selected from the diffusion coefficient table in accordance with a combination of color component values of the target pixel.

Claim 21

The method according to claim 13 , wherein in the error addition step, a diffusion coefficient is selected from the diffusion coefficient table in accordance with a combination of average values, maximum values, or minimum values of color components of the target pixel and a neighboring pixel thereof.

Claim 22

The method according to claim 13 , wherein in the error addition step, a diffusion coefficient is selected from the diffusion coefficient table in accordance with a product of color component values of the target pixel and a neighboring pixel thereof.

Claim 23

An image processing method of quantizing multilevel color image data containing at least two kinds of color components, comprising: an error addition step of adding a quantization error value distributed from a neighboring pixel for each color component to each of a plurality of color components contained in a target pixel; a modulation amount determination step of determining a threshold modulation amount of each color in accordance with a combination of color component values of pixels including the target pixel; and a quantization step of acquiring a threshold for each color component, in accordance with a combination of color components of the target pixel, from a threshold table which stores a quantization threshold for each of the color components, in accordance with a combination of color component values of pixels, determining a threshold modulated by adding the modulation amount to the threshold for each color, and quantizing the target pixel in accordance with a relationship in magnitude between the modulated threshold and a value of each color component to which an error value is added in the error addition step, wherein in the error addition step, an error value is added to the target pixel in accordance with a combination of color components of the target pixel and a diffusion coefficient selected from a diffusion coefficient table which stores a diffusion coefficient for diffusing a quantization error produced in the quantization step in accordance with a combination of color component values of a pixel.

Claim 24

An image processing method of quantizing multilevel color image data containing at least two kinds of color components, comprising: an error addition step of adding a quantization error value distributed from a neighboring pixel for each color component to each of a plurality of color components contained in a target pixel; a modulation amount determination step of determining a threshold modulation amount of each color in accordance with a combination of color component values of pixels including the target pixel; and a quantization step of acquiring a threshold for each color component, in accordance with a combination of color components of the target pixel, from a threshold table which stores a quantization threshold for each of the color components, in accordance with a combination of color component values of pixels, determining a threshold modulated by adding the modulation amount to the threshold for each color, and quantizing the target pixel in accordance with a relationship in magnitude between the modulated threshold and a value of each color component to which an error value is added in the error addition step, wherein in the error addition step, an error value is added to the target pixel in accordance with a combination of color components of the target pixel and a diffusion coefficient selected from a diffusion coefficient table which stores a diffusion coefficient for diffusing a quantization error produced in the quantization step in accordance with a combination of color component values of a pixel.

Claim 25

An image processing method of quantizing multilevel color image data containing at least three kinds of color components, comprising: an error addition step of adding a quantization error value distributed from a neighboring pixel for each color component to each of a plurality of color components contained in a target pixel; a modulation amount determination step of determining a threshold modulation amount of each color in accordance with a combination of color component values of pixels including the target pixel; and a quantization step of acquiring a threshold for each color component, in accordance with a combination of color components of the target pixel, from a threshold table which stores a quantization threshold for each of the color components in accordance with a combination of color component values of pixels, determining a threshold modulated by adding the modulation amount to the threshold for each color, and quantizing the target pixel in accordance with a relationship in magnitude between the modulated threshold and a value of each color component to which an error value is added in the error addition step, wherein a combination of two kinds of color components is quantized by said image processing method, and a remaining color component is quantized with a value of a color component other than a target color component being regarded as 0.

Claim 26

An image processing method of quantizing multilevel color image data containing at least three kinds of color components, comprising: an error addition step of adding a quantization error value distributed from a neighboring pixel for each color component to each of a plurality of color components contained in a target pixel; a modulation amount determination step of determining a threshold modulation amount of each color in accordance with a combination of color component values of pixels including the target pixel; and a quantization step of acquiring a threshold for each color component, in accordance with a combination of color components of the target pixel, from a threshold table which stores a quantization threshold for each of the color components in accordance with a combination of color component values of pixels, determining a threshold modulated by adding the modulation amount to the threshold for each color, and quantizing the target pixel in accordance with a relationship in magnitude between the modulated threshold and a value of each color component to which an error value is added in the error addition step, wherein a combination of two kinds of color components is quantized by said image processing method, and a remaining color component is quantized with a value of a color component other than a target color component being regarded as 0.

Claim 27

A computer-readable medium encoding a program for causing a computer to quantize multilevel color image data containing at least two kinds of color components, said program comprising: a code for an error addition step of adding a quantization error value distributed from a neighboring pixel for each color component to each of a plurality of color components contained in a target pixel; a code for a modulation amount determination step of determining a threshold modulation amount of each color in accordance with a combination of color component values of pixels including the target pixel; and a code for a quantization step of acquiring a threshold for each color component from a threshold table in accordance with a combination of color components of the target pixel, determining a threshold modulated by adding the modulation amount to the threshold for each color, and quantizing the target pixel in accordance with a relationship in magnitude between the modulated threshold and a value of each color component to which an error value is added in the error addition step.

Claim 28

A computer-readable medium encoding a program for causing a computer to quantize multilevel color image data containing at least two kinds of color components, said program comprising: a code for an error addition step of adding a quantization error value distributed from a neighboring pixel for each color component to each of a plurality of color components contained in a target pixel; a code for a modulation amount determination step of determining a threshold modulation amount of each color in accordance with a combination of color component values of pixels including the target pixel; and a code for a quantization step of acquiring a threshold for each color component from a threshold table in accordance with a combination of color components of the target pixel, determining a threshold modulated by adding the modulation amount to the threshold for each color, and quantizing the target pixel in accordance with a relationship in magnitude between the modulated threshold and a value of each color component to which an error value is added in the error addition step.

Claim 29

The medium according to claim 19 , wherein in the modulation amount determination step, a threshold modulation amount for each color is determined in accordance with a combination of color component values of the target pixel.

Claim 30

The medium according to claim 19 , wherein in the modulation amount determination step, a threshold modulation amount for each color is determined in accordance with a combination of average values, maximum values, or minimum values of color components of the target pixel and a neighboring pixel thereof.

Claim 31

A computer-readable medium encoding a program for causing a computer to quantize multilevel color image data containing at least two kinds of color components, said program comprising: a code for an error addition step of adding a quantization error value distributed from a neighboring pixel for each color component to each of a plurality of color components contained in a target pixel; a code for a quantization step of quantizing each color component of the target pixel; and a diffusion coefficient table, wherein in the code for the error addition step, a diffusion coefficient is selected, in accordance with a combination of color components of the target pixel, from a diffusion coefficient table which stores a diffusion coefficient for diffusing the quantization error in accordance with a combination of color component values of a pixel, and an error value is added to the target pixel in accordance with the selected diffusion coefficient.

Claim 32

A computer-readable medium encoding a program for causing a computer to quantize multilevel color image data containing at least two kinds of color components, said program comprising: a code for an error addition step of adding a quantization error value distributed from a neighboring pixel for each color component to each of a plurality of color components contained in a target pixel; a code for a quantization step of quantizing each color component of the target pixel; and a diffusion coefficient table, wherein in the code for the error addition step, a diffusion coefficient is selected, in accordance with a combination of color components of the target pixel, from a diffusion coefficient table which stores a diffusion coefficient for diffusing the quantization error in accordance with a combination of color component values of a pixel, and an error value is added to the target pixel in accordance with the selected diffusion coefficient.

Claim 33

The medium according to claim 22 , wherein in the error addition step, a diffusion coefficient is selected from the diffusion coefficient table in accordance with a combination of color component values of the target pixel.

Claim 34

The medium according to claim 22 , wherein in the error addition step, a diffusion coefficient is selected from the diffusion coefficient table in accordance with a combination of average values, maximum values, or minimum values of color components of the target pixel and a neighboring pixel thereof.

Claim 35

The medium according to claim 22 , wherein in the error addition step, a diffusion coefficient is selected from the diffusion coefficient table in accordance with a product of color component values of the target pixel and a neighboring pixel thereof.

Claim 36

A computer-readable medium encoding a computer program for causing a computer to quantize multilevel color image data containing at least two kinds of color components, said program comprising: a code for an error addition step of adding a quantization error value distributed from a neighboring pixel for each color component to each of a plurality of color components contained in a target pixel; a code for a modulation amount determination step of determining a threshold modulation amount of each color in accordance with a combination of color component values of pixels including the target pixel; and a code for a quantization step of acquiring a threshold for each color component, in accordance with a combination of color components of the target pixel, from a threshold table which stores a quantization threshold for each of the color components in accordance with a combination of color component values of pixels, determining a threshold modulated by adding the modulation amount to the threshold for each color, and quantizing the target pixel in accordance with a relationship in magnitude between the modulated threshold and a value of each color component to which an error value is added in the error addition step, wherein in the code for the error addition step, a diffusion coefficient is selected, in accordance with a combination of color components of the target pixel, from a diffusion coefficient table which stores a diffusion coefficient for diffusing a quantization error in accordance with a combination of color component values of a pixel, and an error value is added to the target pixel in accordance with the selected diffusion coefficient.

Claim 37

A computer-readable medium encoding a computer program for causing a computer to quantize multilevel color image data containing at least two kinds of color components, said program comprising: a code for an error addition step of adding a quantization error value distributed from a neighboring pixel for each color component to each of a plurality of color components contained in a target pixel; a code for a modulation amount determination step of determining a threshold modulation amount of each color in accordance with a combination of color component values of pixels including the target pixel; and a code for a quantization step of acquiring a threshold for each color component, in accordance with a combination of color components of the target pixel, from a threshold table which stores a quantization threshold for each of the color components in accordance with a combination of color component values of pixels, determining a threshold modulated by adding the modulation amount to the threshold for each color, and quantizing the target pixel in accordance with a relationship in magnitude between the modulated threshold and a value of each color component to which an error value is added in the error addition step, wherein in the code for the error addition step, a diffusion coefficient is selected, in accordance with a combination of color components of the target pixel, from a diffusion coefficient table which stores a diffusion coefficient for diffusing a quantization error in accordance with a combination of color component values of a pixel, and an error value is added to the target pixel in accordance with the selected diffusion coefficient.