Main Content

graycoprops

Properties of gray-level co-occurrence matrix (GLCM)

Description

example

stats = graycoprops(glcm,properties) calculates the statistics specified in properties from the gray-level co-occurrence matrix glcm.

graycoprops normalizes the gray-level co-occurrence matrix (GLCM) so that the sum of its elements is equal to 1. Each element (r,c) in the normalized GLCM is the joint probability occurrence of pixel pairs with a defined spatial relationship having gray level values r and c in the image. graycoprops uses the normalized GLCM to calculate properties.

Examples

collapse all

Create simple sample GLCM.

glcm = [0 1 2 3;1 1 2 3;1 0 2 0;0 0 0 3]
glcm = 4×4

     0     1     2     3
     1     1     2     3
     1     0     2     0
     0     0     0     3

Calculate statistical properties of the GLCM.

stats = graycoprops(glcm)
stats = struct with fields:
       Contrast: 2.8947
    Correlation: 0.0783
         Energy: 0.1191
    Homogeneity: 0.5658

Read grayscale image into the workspace.

I = imread('circuit.tif');

Create two gray-level co-occurrence matrices (GLCM) from the image, specifying different offsets.

glcm = graycomatrix(I,'Offset',[2 0;0 2])
glcm = 
glcm(:,:,1) =

       14205        2107         126           0           0           0           0           0
        2242       14052        3555         400           0           0           0           0
         191        3579        7341        1505          37           0           0           0
           0         683        1446        7184        1368           0           0           0
           0           7         116        1502       10256        1124           0           0
           0           0           0           2        1153        1435           0           0
           0           0           0           0           0           0           0           0
           0           0           0           0           0           0           0           0


glcm(:,:,2) =

       13938        2615         204           4           0           0           0           0
        2406       14062        3311         630          23           0           0           0
         145        3184        7371        1650         133           0           0           0
           2         371        1621        6905        1706           0           0           0
           0           0         116        1477        9974        1173           0           0
           0           0           0           1        1161        1417           0           0
           0           0           0           0           0           0           0           0
           0           0           0           0           0           0           0           0

Get statistics on contrast and homogeneity of the image from the GLCMs.

stats = graycoprops(glcm,{'contrast','homogeneity'})
stats = struct with fields:
       Contrast: [0.3420 0.3567]
    Homogeneity: [0.8567 0.8513]

Input Arguments

collapse all

GLCM, specified as one of the following. You can use the graycomatrix function to create a GLCM.

  • An m-by-n matrix of nonnegative integers for a single GLCM

  • An m-by-n-by-p array of nonnegative integers for p valid GLCMs.

Data Types: single | double | int8 | int16 | int32 | int64 | uint8 | uint16 | uint32 | uint64 | logical

Statistical properties of the image derived from GLCM, specified as a comma-separated list string scalars or character vectors, cell array of string scalars or character vectors, string array, space-separated character vector, or "all". You can specify any of the property names listed in this table.

Property

Description

Formula

"Contrast"

Returns a measure of the intensity contrast between a pixel and its neighbor over the whole image.

Range = [0 (size(GLCM,1)-1)^2] 

Contrast is 0 for a constant image.

The property Contrast is also known as variance and inertia.

i,j|ij|2p(i,j)

"Correlation"

Returns a measure of how correlated a pixel is to its neighbor over the whole image.

Range = [-1 1]

Correlation is 1 or -1 for a perfectly positively or negatively correlated image. Correlation is NaN for a constant image.

i,j(iμi)(jμj)p(i,j)σiσj

"Energy"

Returns the sum of squared elements in the GLCM.

Range = [0 1]

Energy is 1 for a constant image.

The property Energy is also known as uniformity, uniformity of energy, and angular second moment.

i,jp(i,j)2

"Homogeneity"

Returns a value that measures the closeness of the distribution of elements in the GLCM to the GLCM diagonal.

Range = [0 1]

Homogeneity is 1 for a diagonal GLCM.

i,jp(i,j)1+|ij|

Example: "Contrast","Homogeneity" specifies two properties as a comma-separated list of string scalars

Example: ["Contrast","Homogeneity"] specifies two properties as an array of string scalars

Example: {"Contrast","Homogeneity"} specifies two properties as cell array of string scalars

Example: 'Contrast Homogeneity' specifies two properties as a space-separated character vector

Data Types: char | string | cell

Output Arguments

collapse all

Statistics derived from the GLCM, returned as a structure with fields that are specified by properties. Each field contains a 1-by-p array, where p is the number of gray-level co-occurrence matrices in glcm. For example, if glcm is an 8-by-8-by-3 array and properties is "Energy", then stats is a structure containing the field Energy, which contains a 1-by-3 array.

Version History

Introduced before R2006a