The document provides a comprehensive overview of image texture, including its definition, types, algorithms, and methods for texture segmentation and classification. It discusses approaches like structured and statistical methods, highlighting tools such as Gabor filters and co-occurrence matrices for analysis. Additionally, the text emphasizes the applications of texture in various fields, including education and technology.

1. TextureTexture
Presented by:
Priyanka Chauhan

2. Contents
Definition
Algorithm
Types
Uses
Texture Segmentation/Classification
Methods used
Approaches
Structured Approach
Statistical Approach
 Use of Texture in education
 Conclusion and References

3. Definition
 An image texture is a set of metrics calculated in
image processing designed to quantify the perceived
texture of an image.
 Regular repetition of an element or pattern on a
surface.
 Image Texture gives us information about the
spatial arrangement of color or intensities in an image
or selected region of an image.
 Texture can be defined as an entity consisting of
mutually related pixels and group of pixels.

4. Definition(contd…)
 In fact, there are many ways in which intensity
might vary, but if the variation does not have
sufficient uniformity, the texture may not be
characterized, sufficiently close to permit recognition
or segmentation.
Thus, the degrees of randomness and of regularity
will have to be measured and compared when
charactering a texture.
Often, textures are derived from tiny objects or
components that are themselves similar, but that are
placed together in ways ranging from purely random
to purely regular, such as bricks in a wall, or grains of
sand, etc.

5. Texture Analysis
Image textures can be artificially created or found
in natural scenes captured in an image.
Since the repetitive local arrangement of intensity
determines the texture, we have to analyze
neighborhoods of pixels to measure texture
properties.
 Because texture has so many different dimensions
no single method of texture representation that is
adequate for a variety of textures.

6. Different textures
(a) (b)
(c)

7. Algorithm
Step1 : Read Image from database x=image.
Step2 : If image in color then convert into grayscales.
Step3 : Apply any one filter on image.
Step4 : If extract region then
Step5 : Display the Segmented Image.
else
Step6 : Repeat step 3.
Step7 : Stored in database.

8. Types of Texture
 Tactile
 Visual
 Natural
 Artificial

9. Tactile Texture
 We perceive it by touching an object
 Every material has a different touch, so it is a
different texture

10. Visual Texture
 When we represent a tactile texture using graphic
elements, we get a visual texture.
 We can create visual textures using different art
procedures: scrapping, stencilling, printing

11. Natural Texture
 They belong to natural elements, like the skin of an
elephant, the cortex of a tree or the surface of a rose
Patel.

12. Artificial Texture
 The surface of any object made by us: steel, paper,
different kinds of fabrics.
 Artificial textures are created and designed by
human beings for a specific purpose to give a sense
of volume.

13. Why do we use texture?
To identify different textured and nontextured
regions in an image.
To classify/segment different texture regions in an
image.
To extract boundaries between major texture
regions.
Analyze texture in CGAnalyze texture in CG
Structured approach Statistical approach

14. Texture Segmentation
 Image texture can be used as a description for
regions into segments
 Can be supervised or unsupervised depending on if
prior knowledge about the image or texture class is
available.
Supervised-identifies and separates one or more
regions that match texture properties shown in the
training textures.
Unsupervised-first recover different texture classes
from an image before separating them into regions.

15. Methods Used for Texture Segmentation
Gabor filter
Edge detection
Thresholding
Histogram based
Region based

16. Gabor Filter
It is a linear filter used for edge detection.
Frequency and orientation representations are
similar to those of the human visual system.
Appropriate for texture representation and
discrimination.
A two dimensional Gabor function g(x, y) is
defined as:

17. Histogram Matching
Compute the histogram of the template.
Sweep a window over the image.
Compute the histogram of the window.
Do a correlation between the histograms.
the texture we
are searching
(template)
window at step k
(thr sample)
window at k+1

18. Structured approach
Structural approach: a set of texels in some regular
or repeated pattern
Repeated 12 times Repeated 12 times

19. Texel
 Can be called as texture element, texel pixel
 Fundamental unit of texture space used in computer
graphics
 Textures are represented by array of texels, just as
pictures are represented by arrays of pixels

20. Statistical approach
 Texture Is a spatial
property.
 A simple one-
dimensional Histogram
is not useful in
characterizing texture.
Example
( an image in which pixels
Alternate From black to
white in a checkerboard
fashion will have the same
histogram as an image in
which the top half is black
and the bottom half is
white).

21. Types of Statistical Approach
Co occurrence matrix
 The gray level co-occurrence matrix approach is
based on studies of the statistics of pixel intensity
distributions.
 The co-occurrence matrices express the relative
frequencies (or probabilities) P(i, j | d,θ) with which
two pixels having relative polar coordinates (d,θ)
appear with intensities i, j.
The co-occurrence matrices provide raw numerical
data on the texture, although this data must be
condensed to relatively few numbers before it can be
used to classify the texture.

22. Co occurrence Matrix
To compute such a matrix, we first separate the
intensity in the image into a small number of
different levels.
For example, by dividing the usual brightness
values ranging from 0 to 255 by 64, we create the
levels 0, 1, 2, and 3.

23. Co occurrence Matrix
Example (2 gray levels):
001111001111
001100001100
001111001111
001100001100
001111001111
001100001100
local texture
patch
dd = (1, 1)= (1, 1)
displacement
vector
1/25 *1/25 *
co-occurrence
matrix
00
11
00 11
22 99
1010 44

24. Edge Detection
 The use of edge detection to determine the number
of edge pixels in a specified region helps determine
a characteristic of texture complexity.
 Consider a region with N pixels. the gradient-based
edge detector is applied to this region by producing
two outputs for each pixel p: the gradient magnitude
Mag(p) and the gradient direction Dir(p) then –
The edgeness per unit area can be defined by
F_{edgeness}={|{p | Mag(p) > T}|}/{N} for some
threshold T.

25. Edge Detection
To include orientation with edgeness we can use
histograms for both gradient magnitude and gradient
direction. Let Hmag(R) denote the normalized
histogram of gradient magnitudes of region R, and
let Hdir denote the normalized histogram of gradient
orientations of region R. Both are normalized
according to the size NRThen is quantitative texture
description of region R.

26. Laws Texture Energy Measures
 Another approach to generate texture features is to
use local masks to detect various types of textures.
Convolution masks of 5x5 are used to compute the
energy of texture .
Law’s measures use a set of convolution filters to
assess gray level, edges, spots, ripples, and waves
in textures.
This method starts with three basic filters:
 averaging: L3 = (1, 2, 1)
 first derivative (edges): E3 = (-1, 0, 1)
 second derivative (curvature): S3 = (-1, 2, -1)

27. Laws Texture Energy Measures
Convolving these filters with themselves and each
other results in five new filters.
The masks are generated from the following
vectors.
 L5 = [ +1 +4 6 +4 +1 ] (Level)
 E5 = [ -1 -2 0 +2 +1 ] (Edge)
 S5 = [ -1 0 2 0 -1 ] (Spot)
 W5 = [ -1 +2 0 -2 +1 ] (Wave)
 R5 = [ +1 -4 6 -4 +1 ] (Ripple)

28. Laws Texture Energy Measures
Now we can multiply any two of these vectors, using the
first one as a column vector and the second one as a row
vector, resulting in 5 × 5 Law’s masks.
For example
L5*S5=
Now you can apply the resulting 25 convolution filters to a
given image.
The 25 resulting values at each position in the image are
useful descriptors of the local texture.

29. Use of Texture in education
 A texture analyzer can be used to accurately, and
repeatable, test the product to measure numerical
output against the subjective findings.
 Food Technology Corporation's texture
measurement systems are in use in schools,
colleges and universities worldwide.
 Other uses are in cosmetics, web designing and
auto parts designing etc.

30. Conclusion and References
 Imagine the world without texture?
https://www.cs.auckland.ac.nz/~georgy/research/text
https://en.wikipedia.org/wiki/Image_texture
 http://ieeexplore.ieee.org/xpl/articleDetails.jsp

31. THANK YOUTHANK YOU