imapprox

Approximate indexed image by reducing number of colors

Syntax

[Y,newmap] = imapprox(X,map,Q)
[Y,newmap] = imapprox(X,map,tol)
Y = imapprox(X,map,inmap)
___ = imapprox(___,dithering)

Description

example

[Y,newmap] = imapprox(X,map,Q) approximates the colors in indexed image X and associated colormap map by using minimum variance quantization with Q quantized colors. imapprox returns the indexed image Y with colormap newmap.

[Y,newmap] = imapprox(X,map,tol) approximates the colors in indexed image X and associated colormap map by using uniform quantization with tolerance tol.

Y = imapprox(X,map,inmap) approximates the colors in indexed image X and associated colormap map by using inverse colormap mapping with colormap inmap. The inverse colormap algorithm finds the colors in inmap that best match the colors in map.

___ = imapprox(___,dithering) enables or disables dithering.

Examples

collapse all

Load an indexed image of a mandrill's face. Display image X using its associated colormap, map, which has 220 colors.

figure
load mandrill
image(X)
colormap(map)
axis off         
axis image

Reduce the number of colors in the indexed image from 220 to only 16 colors by producing a new image, Y, and its associated colormap, newmap.

figure
[Y,newmap] = imapprox(X,map,16);
image(Y)
colormap(newmap)
axis off       
axis image

Input Arguments

collapse all

Indexed image with many colors, specified as an m-by-n matrix of nonnegative integers.

Data Types: single | double | uint8 | uint16

Colormap with many colors associated with indexed image X, specified as a d-by-3 matrix with values in the range [0, 1]. Each row of map is a three-element RGB triplet that specifies the red, green, and blue components of a single color of the colormap.

Data Types: double

Number of quantized colors used for minimum variance quantization, specified as a positive integer that is less than or equal to 65,536. The returned colormap newmap has Q or fewer colors.

Tolerance used for uniform quantization, specified as a number in the range [0, 1]. The returned colormap newmap has (floor(1/tol)+1)^3 or fewer colors.

Colormap with fewer colors used for inverse colormap mapping, specified as a c-by-3 matrix with values in the range [0, 1]. Each row of inmap is a three-element RGB triplet that specifies the red, green, and blue components of a single color of the colormap. The colormap has a maximum of 65,536 colors.

Data Types: double

Perform dithering, specified as 'dither' or 'nodither'. Dithering increases the color resolution at the expense of spatial resolution. For more information, see dither.

If you select 'nodither', then imapprox does not perform dithering. Instead, the function maps each color in the original image to the closest color in the new colormap.

Output Arguments

collapse all

Indexed image with fewer colors, returned as an m-by-n matrix of positive integers. If the length of newmap (or immap, if specified) is less than or equal to 256, then the output image is of class uint8. Otherwise, the output image is of class double.

Data Types: double | uint8

Colormap with fewer colors associated with the output indexed image Y, returned as a c-by-3 matrix with values in the range [0, 1]. Each row of newmap is a three-element RGB triplet that specifies the red, green, and blue components of a single color of the colormap. The colormap has a maximum of 65,536 colors.

Data Types: double

Algorithms

imapprox uses rgb2ind to create a new colormap that uses fewer colors. For more information about quantization and inverse colormap mapping, see the Algorithms of rgb2ind.

Introduced before R2006a