Scribd
Upload
66 views19 pages

Unit 3 Session 1

The document discusses feature extraction methods for categorizing textures in images. It describes extracting texture features from training images to build models for each texture. Common feature extraction methods include statistical, structural, model-based, and transform-based approaches. These features are then used to classify textures in test images using machine learning classification algorithms. The classification accuracy is evaluated by comparing estimated classes to actual classes.

Uploaded by

Ankur Jha
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
66 views19 pages

Unit 3 Session 1

The document discusses feature extraction methods for categorizing textures in images. It describes extracting texture features from training images to build models for each texture. Common feature extraction methods include statistical, structural, model-based, and transform-based approaches. These features are then used to classify textures in test images using machine learning classification algorithms. The classification accuracy is evaluated by comparing estimated classes to actual classes.

Uploaded by

Ankur Jha
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 19

18ECE340T- MACHINE PERCEPTION WITH COGNITION

UNIT-3

Representing Textures

Reference Book
Computer and Machine Vision by E.R.Davies
Texture
Two kinds of texture : Tactile and Optical.
◦ where we can feel the tactile texture by touching or seeing the surface.
◦ while the e optical or visual texture, refers to the shape and content of the image. Humans can easily
diagnose the texture of the image but making a machine to analyze the texture of the image has its
complexity.
◦ In the field of image processing, we can consider the spatial changes of the brightness intensity of the
pixel as the texture of the image. 
In image processing, textural images are those images in which a specific pattern of texture distribution
is repeated sequentially throughout the image. Thabovew image can be a representation of the textural
image where part (b), represents the repeated pattern on the texture.
Texture
Texture
Analysis
Applications :
1.Face Recognition
Texture Texture
Segmentation Synthesis 2.Tracking objects in the videos
3.Diagnosis of product quality
Texture
Texture Shape
Claasification 4.Medical image analysis
5.Remote sensing
Statistical Mode based
Methods Methods 6.Vegetation

Structural Transform
Methods based methids
Texture
Texture
Representing Textures
• Texture is a phenomenon that is widespread.
• It is easy to recognize, and hard to define.
• The effect is referred to as texture or not depends on the scale at which it is viewed.
• Texture was defined as the characteristic variation in intensity of a region of an
image that should allow us to recognize and describe it and to outline its boundaries.
• Ex: A leaf that occupies most of an image is an object, but the foliage of a tree is a
texture.
• Views of large numbers of small objects are often best thought of as textures.
• Examples include grass, foliage, brush, pebbles, and hair.
Texture Types
• Definition of Texture

• Texture types

Synthetic Natural Stochastic


< Prev Next >
Texture Definition
• Texture: The regular repetition of an element or pattern on a surface.
• Purpose of texture analysis:
– To identify different textured and non- textured regions
in image.
– To classify/segment different texture regions in an image.
• To extract boundaries between major texture regions
– To describe the texel unit.
– 3-D shape from texture
Texture

• Texture is actually a very unclear concept, often


attributed to human perception, as either the feel or
the appearance of (woven) fabric.
• Everyone has their own interpretation as to the
nature of texture; there is no mathematical
definition for texture.
• In textile --🡪arrangement of threads.
• characteristic feel - arrangement of small
constituent parts, perceived structure, (of skin,
rock,soil, organic tissue, literary work, etc.)
• Representation of structure and detail of objects in
art .
Texture
• Texture depends not only on the input image, but also on the goal for which the
image texture is used and the textural features that are extracted from the image.
• Essentially, there is no unique definition of texture. There is no unique
mathematical model to synthesis texture; there are many ways to describe and
extract it.
• Clearly, images will usually contain samples of more than one texture.
• To be able to describe texture (texture descriptions are measurements which
characterize a texture) and then to classify it (classification is attributing the
correct class label to a set of measurements) and then perhaps to segment an
image according to its texture content.
Texture Characterization
• The texels will have various sizes and degrees of uniformity.
• The texels will be orientated in various directions.
• The texels will be spaced at varying distances in different directions.
• The contrast will have various magnitudes and variations.
• Various amounts of background may be visible between texels.
• The variations composing the texture may each have varying degrees of regularity face to
face randomness.
Texture Representation
There are three main kinds of texture representation.
1.Local texture representations
2.pooled texture representations.
3.Data-driven texture representations

Local texture representations


• Local texture representations encode the texture very close to a point in an image.
• These representations can’t be comprehensive, because they look at a small piece of the image.
• It is very useful in image segmentation. Inside a region look similar to one another, and different from points
outside the region, and segmentation algorithms need a description of the appearance close to the point to
impose this requirement.
• Recovering surface orientation or surface shape from an image texture is known as shape from texture.
Texture Representation
Pooled Texture Representation:
A texture is a set of textons that repeat in some way. Find these textons by looking for image patches that are
common.
• An alternative is to find sets of texton sub-elements—that is, vectors of filter outputs—that are common (if
textons are repeated, then so are their subelements).
There are two important difficulties in finding image patches or vectors of filter outputs that commonly occur
together.
(i) Representations of the image are continuous, cannot simply count how many times a particular pattern
occurs, because each vector is slightly different.
(ii) The representation is high dimensional in either case. A patch around a pixel might need hundreds of
pixels to represent it well; similarly, hundreds of different filters might be needed to represent the image at a pixel.
• This means we cannot build a histogram directly, either, because it will have an unmanageable number of cells.
Texture Representation
Data-driven Texture Representations

• Data-driven texture representations model a texture by a procedure that can generate a textured region from
an example.
• These representations are not appropriate for segmentation or recognition applications, but are tremendously
valuable for texture synthesis.
• The texture on a surface can be a strong cue to its shape. If the texture is “the same” over the surface, then
deformation of the texture from point to point can be a cue to the shape of the surface.
• For example, if we have a perspective view of an inclined plane with spots on it, the spots will be smaller
closer to the horizon in the image. This can be used to recover the inclination of the plane.
• Similarly, on a curved surface, the foreshortening of texture elements gives some information about the local
inclination of the surface.
• Recovering surface orientation or surface shape from an image texture is known as shape from texture;
Challenges in Texture Analysis
❖Rotation image
❖Noise image

Also, there can be various chances of images to differ from each other in terms of
scale, viewpoint, brightness or intensity of light. – which causes challenges in
texture classification.
Feature Extraction Method for Categorizing Textures
Feature Extraction Method for Categorizing Textures
The feature extraction part:
Try to extract the textual properties of the images
The motive of this part is to make a model which can deal with every texture of the image that exists
in training time.
The classification part:
In this part, we perform the texture analysis on the test images with the same techniques which we
applied for the training images and apply a classification algorithm which can be a statistic or deep
learning algorithm.

The images get examined by the feature extractor and then texture classification is done by the
classification algorithm
Feature Extraction Method for Categorizing Textures
Feature Extraction
The basic idea is to extract texture features from the images, and for this procedure, it is required to have a model
for every texture available in the training images.
These features can be discrete histogram, numerical, empirical distribution, and texture features such as contrast,
spatial structure, direction, etc.
The extracted texture feature can be used for teaching classification. There can be various ways to classify
texture and the efficiency of these ways can be dependent on the type of texture features extracted. These
methods can be divided into the following groups:
◦ Statistical methods
◦ Structural methods
◦ Model-based methods
◦ Transformer methods
We can use any of the methods for extracting features from the images.
Feature Extraction Method for Categorizing Textures
Classification
In the second stage of the process, classification is performed on the extracted texture features based on the
machine learning algorithms with classification algorithms.
Using the classification method, appropriate classes for each texture are selected.
Using the comparison between the vector of the extracted texture feature from the extraction part of the
process and the vector of the selection test phase characteristics, classes are determined.
This step is repeated for every image presented.
The estimated classes for evaluation with the actual class are adapted and the recognition rate is calculated
which shows the efficiency of the implemented algorithm. Normally applied accuracy measure is:
Classification accuracy = (correct matches / number of test image) ×100

You might also like