Main Content

adapthisteq

Contrast-limited adaptive histogram equalization (CLAHE)

Description

example

J = adapthisteq(I) enhances the contrast of the grayscale image I by transforming the values using contrast-limited adaptive histogram equalization (CLAHE) [1].

J = adapthisteq(I,Name,Value) uses name-value arguments to control aspects of the contrast enhancement.

Examples

collapse all

Apply CLAHE to an image and display the results.

I = imread('tire.tif');
J = adapthisteq(I,'clipLimit',0.02,'Distribution','rayleigh');
imshowpair(I,J,'montage');
title('Original Image (left) and Contrast Enhanced Image (right)')

Figure contains an axes object. The axes object with title Original Image (left) and Contrast Enhanced Image (right) contains an object of type image.

Read the indexed color image into the workspace.

[X, MAP] = imread('shadow.tif');

Convert the indexed image into a truecolor (RGB) image, then convert the RGB image into the L*a*b* color space.

RGB = ind2rgb(X,MAP);
LAB = rgb2lab(RGB);

Scale values to the range expected by the adapthisteq function, [0 1].

L = LAB(:,:,1)/100;

Perform CLAHE on the L channel. Scale the result to get back to the range used by the L*a*b* color space.

L = adapthisteq(L,'NumTiles',[8 8],'ClipLimit',0.005);
LAB(:,:,1) = L*100;

Convert the resulting image back into the RGB color space.

J = lab2rgb(LAB);

Display the original image and the processed image.

figure
imshowpair(RGB,J,'montage')
title('Original (left) and Contrast Enhanced (right) Image')

Figure contains an axes object. The axes object with title Original (left) and Contrast Enhanced (right) Image contains an object of type image.

Shadows in the enhanced image look darker and highlights look brighter. The overall contrast is improved.

Input Arguments

collapse all

Grayscale image, specified as a 2-D numeric matrix.

Data Types: single | double | int16 | uint8 | uint16

Name-Value Arguments

Specify optional pairs of arguments as Name1=Value1,...,NameN=ValueN, where Name is the argument name and Value is the corresponding value. Name-value arguments must appear after other arguments, but the order of the pairs does not matter.

Before R2021a, use commas to separate each name and value, and enclose Name in quotes.

Example: "NumTiles",[8 16] divides the image into 8 rows and 16 columns of tiles.

Number of rectangular contextual regions (tiles) into which adapthisteq divides the image, specified as a 2-element vector of positive integers. With the original image divided into M rows and N columns of tiles, the value of NumTiles is [M N]. Both M and N must be at least 2. The total number of tiles is equal to M*N. The optimal number of tiles depends on the type of the input image, and it is best determined through experimentation.

Data Types: double

Contrast enhancement limit, specified as a number in the range [0, 1]. Higher limits result in more contrast.

ClipLimit is a contrast factor that prevents oversaturation of the image specifically in homogeneous areas. These areas are characterized by a high peak in the histogram of the particular image tile due to many pixels falling inside the same gray level range. Without the clip limit, the adaptive histogram equalization technique could produce results that, in some cases, are worse than the original image.

Data Types: double

Number of histogram bins used to build a contrast enhancing transformation, specified as a positive integer. Higher values result in greater dynamic range at the cost of slower processing speed.

Data Types: double

Range of the output image data, specified as one of these values.

ValueDescription
"full"Use the full range of the output class (such as [0 255] for uint8).
"original"Limit the range to [min(I(:)) max(I(:))].

Data Types: char | string

Desired histogram shape, specified as one of the following values:

ValueDescription
"uniform"Create a flat histogram.
"rayleigh"Create a bell-shaped histogram.
"exponential"Create a curved histogram.

Distribution specifies the distribution that adapthisteq uses as the basis for creating the contrast transform function. The distribution you select should depend on the type of the input image. For example, underwater imagery appears to look more natural when the Rayleigh distribution is used.

Data Types: char | string

Distribution parameter, specified as a nonnegative number. Alpha is only used when Distribution is set to "rayleigh" or "exponential".

Data Types: double

Output Arguments

collapse all

Contrast enhanced image, returned as a 2-D matrix of the same data type as the input image I.

Algorithms

CLAHE operates on small regions in the image, called tiles, rather than the entire image. adapthisteq calculates the contrast transform function for each tile individually. Each tile's contrast is enhanced, so that the histogram of the output region approximately matches the histogram specified by the Distribution value. The neighboring tiles are then combined using bilinear interpolation to eliminate artificially induced boundaries. The contrast, especially in homogeneous areas, can be limited to avoid amplifying any noise that might be present in the image.

References

[1] Zuiderveld, Karel. "Contrast Limited Adaptive Histograph Equalization." Graphic Gems IV. San Diego: Academic Press Professional, 1994. 474–485.

Extended Capabilities

GPU Code Generation
Generate CUDA® code for NVIDIA® GPUs using GPU Coder™.

Version History

Introduced before R2006a

expand all