Visual Texture is now seen as the critical statistical measure for the description of images for the query by image content of the massive image databases now emerging.
Visual Texture is a concept that is very hard to define. Images of man-made textures such as woven cloth, raffia are a classic example of quasi-uniform textures. The 17 mathematically precise patterns, technically called wallpaper designs, are one extreme form of (uniform) visual texture. But so are images of leather, of sand, of clouds, and even (arguably) spider's webs. Regions of common visual texture have at a certain scale statistical parameters that are (nearly) equal; eg, mean gray-scale, and its local variance.

TEXTURE INDEX An index to visual texture within image engineering
What is Visual Texture?
Texture Recognition Segmentation
Texture Classification League Table
Texture Standards
Useful Links and References

R.L.Kaskyap and Y. Choe, Multi-level 3D Rotational Invariant Classification Proc 11th IAPR Conf, The Hague, nl, Aug 30-Sep 3, 1992, Vol II, pp 251-255.
Harvey A. Cohen and You Jia,A multi-scale texture classifier based on multi-resolution "tuned" masks, Pattern Recognition Letters, Vol 13 No 8 August 1992, 599-603
Jane You and Harvey A. Cohen, Classification and Segmentation of Rotated and Scaled Textured Images Using Texture Tuned Masks, International Journal of Pattern Recognition, Vol 26, No 2, pp 245-258, 1993.
Harvey A. Cohen ( with Jane You and Hong Shen of Griffith University) An Efficient Parallel Texture Classification for Image Retrieval Journal of Visual Language and Computing, Vol 8 No 3 June 1997, pp 359-372.
M. Pietikainen, Azriel Rosenfeld and Larry S. Davis, Experiments with Texture Classification Using Averages of Local Pattern Matches IEEE Systems, Man, Cybernetics, SMC-13, pp 421-426, 1983.
F.S. Cohen, Z. Fan, M.A. Patel, Classification of Rotated and Scaled Textured Images Using Gaussian Markov Random Field Models IEEE PAMI Vol 13 No 2 February 1991, pp 192-201.
Texture Modelling at MIT Media Lab.

Classic standard for texture classification has been the 112 images in the album designed for use by graphic artists, Phil Brodatz, Textures: A Phographic Album for Artists and Designers Dover Publications, New York, 1966. Some digitised versions are available by courtesy of Trygve Randen The Brodatz Textures used in experiments on texture classifying are tabled below:
    Name Brodatz
    FS Cohen
    et al
    Calf D24 x
    Canvas D20 x
    Canvas D21 x
    Cork D4 x
    Herringbone D17 x
    Paper D57 x x
    Pebbles D54x
    Pigskin D92 x
    Raffia D84 x x
    Sand D29 x x
    Straw D15 x
    Wire D14 x
    Wire D6 x
    Water D38 x
    Wood D68 x
    Wool D19 x x
There have been rumours that copyright issues with regard to the use of digital images derived from the Brodatz album have recently arisen. This, if true, is unfortunate, as there is need for on pen standard so researchers can adequtely compare different approaches to texture classification. Currently Roz Picard et al at the Media Lab, MIT, are preparing a VisTex Data Base a compilation of 100+ visual textures freely useable by researchers.

Image Spider WWW Links
Links to Interesting Image Engineering Sites

Image Engineering at La Trobe University

This section will grow very soon.