Linux Certif

Toute la documentation sur la certification Linux LPI

r.texture.1grass

NAME

r.texture - Generate images with textural features from a raster map.

KEYWORDS

raster

SYNOPSIS

r.texture
r.texture help
r.texture [-qackviswxedpmno] input=name prefix=string [size=value] [distance=value] [--overwrite] [--verbose] [--quiet]

Flags:

-q

Quiet

-a

Angular Second Moment

-c

Contrast

-k

Correlation

-v

Variance

-i

Inverse Diff Moment

-s

Sum Average

-w

Sum Variance

-x

Sum Entropy

-e

Entropy

-d

Difference Variance

-p

Difference Entropy

-m

Measure of Correlation-1

-n

Measure of Correlation-2

-o

Max Correlation Coeff

--overwrite

Allow output files to overwrite existing files

--verbose

Verbose module output

--quiet

Quiet module output

Parameters:

input=name

Name of input raster map

prefix=string

Prefix for output raster map(s)

size=value

The size of sliding window (odd and >= 3)
Default: 3

distance=value

The distance between two samples (>= 1)
Default: 1

DESCRIPTION

r.texture - Creates map raster with textural features for user-specified raster map layer. The module calculates textural features based on spatial dependence matrices at 0, 45, 90, and 135 degrees for a distance (default = 1).

In general, several variables constitute texture: differences in grey level values, coarseness as scale of grey level differences, presence or lack of directionality and regular patterns.

r.texture reads a GRASS raster map as input and calculates textural features based on spatial dependence matrices for north-south, east-west, northwest, and southwest directions using a side by side neighborhood (i.e., a distance of 1). Be sure to carefully set your resolution (using g.region) before running this program, or else your computer could run out of memory. Also, make sure that your raster map has no more than 255 categories. The output consists into four images for each textural feature, one for every direction.

A commonly used texture model is based on the so-called grey level co-occurrence matrix. This matrix is a two-dimensional histogram of grey levels for a pair of pixels which are separated by a fixed spatial relationship. The matrix approximates the joint probability distribution of a pair of pixels. Several texture measures are directly computed from the grey level co-occurrence matrix.

The following are brief explanations of texture measures:

 Angular Second Moment (ASM, also called Uniformity): This is a measure of local homogeneity and the opposite of Entropy. High values of ASM occur when the pixels in the moving window are very similar.
Note: The square root of the ASM is sometimes used as a texture measure, and is called Energy.

 Inverse Difference Moment (IDM, also called Homogeneity): This measure relates inversely to the contrast measure. It is a direct measure of the local homogeneity of a digital image. Low values are associated with low homogeneity and vice versa.

 Contrast (Contr): This measure analyses the image contrast (locally gray-level variations) as the linear dependency of grey levels of neighboring pixels (similarity). Typically high, when the scale of local texture is larger than the distance.

 Correlation (Corr): This measure analyses the linear dependency of grey levels of neighboring pixels. Typically high, when the scale of local texture is larger than the distance.

 Variance (Var): A measure of gray tone variance within the moving window (second-order moment about the mean)

 Difference Variance (DV): ...

 Sum Variance (SV): ... 

 Sum Average (SA): ...

 Entropy (Entr): This measure analyses the randomness. It is high when the values of the moving window have similar values. It is low when the values are close to either 0 or 1 (i.e. when the pixels in the local window are uniform).

 Difference Entropy (DE): ...

 Sum Entropy (SE): ...

 Information Measures of Correlation (MOC): ...

 Maximal Correlation Coefficient (MCC): ...

NOTES

Algorithm taken from:
Haralick, R.M., K. Shanmugam, and I. Dinstein. 1973. Textural features for image classification. IEEE Transactions on Systems, Man, and Cybernetics, SMC-3(6):610-621.

The code was taken by permission from pgmtexture, part of PBMPLUS (Copyright 1991, Jef Poskanser and Texas Agricultural Experiment Station, employer for hire of James Darrell McCauley).
Man page of pgmtexture

EXAMPLE

Calculation of Angular Second Moment of B/W orthophoto (North Carolina data set):
g.region rast=ortho_2001_t792_1m -p
r.texture -a ortho_2001_t792_1m prefix=ortho_texture

# display
g.region n=221461 s=221094 w=638279 e=638694
d.shadedmap drape=ortho_texture_ASM_0 rel=ortho_2001_t792_1m
This calculates four maps (requested texture at four orientations): ortho_texture_ASM_0, ortho_texture_ASM_45, ortho_texture_ASM_90, ortho_texture_ASM_135.

BUGS

- The program can run incredibly slow for large raster maps.

- The method for finding the maximal correlation coefficient, which requires finding the second largest eigenvalue of a matrix Q, does not always converge.

REFERENCES

Haralick, R.M., K. Shanmugam, and I. Dinstein (1973). Textural features for image classification. IEEE Transactions on Systems, Man, and Cybernetics, SMC-3(6):610-621.

Bouman, C. A., Shapiro, M.,(March 1994).A Multiscale Random Field Model for Bayesian Image Segmentation, IEEE Trans. on Image Processing, vol. 3, no.2.

Haralick, R., (May 1979). Statistical and structural approaches to texture, Proceedings of the IEEE, vol. 67, No.5, pp. 786-804

Hall-Beyer, M. (2007). The GLCM Tutorial Home Page (Grey-Level Co-occurrence Matrix texture measurements). University of Calgary, Canada

SEE ALSO

i.smap, i.gensigset, i.pca, r.digit, i.group

AUTHOR

G. Antoniol - RCOST (Research Centre on Software Technology - Viale Traiano - 82100 Benevento)
C. Basco - RCOST (Research Centre on Software Technology - Viale Traiano - 82100 Benevento)
M. Ceccarelli - Facolta di Scienze, Universita del Sannio, Benevento

Last changed: $Date: 2008-11-02 08:51:16 +0100 (dom, 02 nov 2008) $

Full index

© 2003-2010 GRASS Development Team