ee.Image.glcmTexture

  • Computes 18 texture metrics, derived from Gray-Level Co-occurrence Matrices (GLCMs), for each band of an image to quantify texture.

  • GLCMs analyze the frequency of pixel brightness value combinations within a specified neighborhood, offering insights into image texture.

  • Users can customize the neighborhood size and kernel to control the analysis scale and direction, with directional averaging available for output simplification.

  • The output includes Haralick's 14 GLCM metrics and Conners' 4 additional metrics, providing a comprehensive texture profile.

  • The provided links to scientific publications offer further details on the underlying theory and calculations of the texture metrics.

 Computes texture metrics from the Gray Level Co-occurrence Matrix around each pixel of every band. The GLCM is a tabulation of how often different combinations of pixel brightness values (grey levels) occur in an image. It counts the number of times a pixel of value X lies next to a pixel of value Y, in a particular direction and distance. and then derives statistics from this tabulation.

This implementation computes the 14 GLCM metrics proposed by Haralick, and 4 additional metrics from Conners. Inputs are required to be integer valued.

The output consists of 18 bands per input band if directional averaging is on and 18 bands per directional pair in the kernel, if not:

ASM: f1, Angular Second Moment; measures the number of repeated pairs

CONTRAST: f2, Contrast; measures the local contrast of an image

CORR: f3, Correlation; measures the correlation between pairs of pixels

VAR: f4, Variance; measures how spread out the distribution of gray-levels is

IDM: f5, Inverse Difference Moment; measures the homogeneity

SAVG: f6, Sum Average

SVAR: f7, Sum Variance

SENT: f8, Sum Entropy

ENT: f9, Entropy. Measures the randomness of a gray-level distribution

DVAR: f10, Difference variance

DENT: f11, Difference entropy

IMCORR1: f12, Information Measure of Corr. 1

IMCORR2: f13, Information Measure of Corr. 2

MAXCORR: f14, Max Corr. Coefficient. (not computed)

DISS: Dissimilarity

INERTIA: Inertia

SHADE: Cluster Shade

PROM: Cluster prominence

More information can be found in the two papers: Haralick et. al, 'Textural Features for Image Classification', https://doi.org/10.1109/TSMC.1973.4309314 and Conners, et al, Segmentation of a high-resolution urban scene using texture operators', https://sdoi.org/10.1016/0734-189X(84)90197-X.

UsageReturns
Image.glcmTexture(size, kernel, average)Image
ArgumentTypeDetails
this: imageImageThe image for which to compute texture metrics.
sizeInteger, default: 1The size of the neighborhood to include in each GLCM.
kernelKernel, default: nullA kernel specifying the x and y offsets over which to compute the GLCMs. A GLCM is computed for each pixel in the kernel that is non-zero, except the center pixel and as long as a GLCM hasn't already been computed for the same direction and distance. For example, if either or both of the east and west pixels are set, only 1 (horizontal) GLCM is computed. Kernels are scanned from left to right and top to bottom. The default is a 3x3 square, resulting in 4 GLCMs with the offsets (-1, -1), (0, -1), (1, -1) and (-1, 0).
averageBoolean, default: trueIf true, the directional bands for each metric are averaged.