Presentation on cbir

8/6/2019 Presentation on cbir

1/33

Fourth Year Project - Presentation

Project Title:

Content Based Image Retrieval (CBIR

)

Presenters:

Rami Al Tayeche

Ahmed KhalilSupervisor:

Professor Aysegul Cuhadar


2/33

Presentation - Outline

Introduction

What is CBIR?

Applications of CBIR

Our Approach

Colour

TextureShape

Where We Are

Conclusion

Questions and Answers


3/33

Introduction - What is CBIR?

The term [CBIR] describes the process of

retrieving desired images from a large collection on

the basis of features (such as colour, texture and

shape) that can be automatically extracted from the

images themselves.


4/33

Introduction - Reasons for its

development

In many current applications with large image

databases, traditional methods of image indexing

have proven to be insufficient.

For example;

Finger print scanning

cannot be done using a

keyword search.


5/33

Introduction - Applications

Automatic face recognition systems


6/33


Medical Image Databases


7/33


TrademarkImage Registration


8/33

Our Approach - Image Features

The image features that we will be focusing on, for

image retrieval are:

Colour

Texture

Shape

Spatial location

Pixel intensity

Other primitive features not considered are:


9/33

Our Approach - Colour


10/33

Our Approach - Colour Histograms


11/33

Our Approach - Colour Maps


12/33

Our Approach - Minkowski Distance


13/33

Our Approach - Quadratic Distance


14/33

Our Approach - Similarity Matrix


15/33

Our Approach - Implementation

Matlab Code


16/33

Our Approach - Texture

o Texture is that innate property of all surfaces that

describes visual patters, and that contain

important information about the structural

arrangement of the surface and its relationship to

the surrounding environment.

What is Texture?


17/33

Our Approach - Texture

Examples:

BrickTexture

Finger print

TextureCloudsTexture

RocksTexture


18/33

Our Approach - Texture Properties

Co-occurrence matrix:

o Based on the orientation and distance between

image pixels.

o From it we obtain statistics that represent:

Coarseness

Contrast

Directionality

Linelikeness

Regularity

Roughness

Texture

properties


19/33

Our Approach - Wavelet Texture

Wavelet Texture:

o Textures can be modeled as quasi-periodic patterns

with spatial/frequency representation.

The wavelet transform

transforms the image

into a multi-scale

representation with both

spatial and frequency

characteristics.


20/33

Our Approach - Tree Algorithm

Algorithm: Tree-Structured WaveletTransform

1. Decompose the image into four sub-images

2. Calculate the energy of all decomposed images at

the same scale, using:

3. If the energy of a sub-image is significantly larger,

repeat from step1.

! !

!

m

i

n

j

jiXMN

E1 1

,1


21/33

Our Approach - Tree Algorithm


22/33

Our Approach - 1st Decomposition


23/33

Our Approach - 2nd Decomposition


24/33

Our Approach - Classification

Algorithm: Euclidean Distance Classification

1. Decompose query image.

2. Get the energies of the first dominant kchannels.

3. For image iin the database obtain the kenergies.

4. Calculate the Euclidean distance between the two

sets of energies, using:

5. Increment i. Repeat from step 3.

!

!k

k

kiki yxD1

2

,


25/33

Our Approach - Shape

o Shape is the characteristic surface configuration

that outlines an object giving it a definite

distinctive form.

o Fairly well-defined concept.

What is Shape?


26/33

Our Approach - Shape

Examples:


27/33

Our Approach - Shape Features

Aspect ratio

Circularity

Moment invariants

Sets of consecutive

boundary segments


28/33

Our Approach - Shape Extraction

Techniques under consideration:

oFourier Descriptor

oMoment Invariants

oDirectional Histograms


29/33

Where We Are


30/33

Image Database


31/33

Conclusion

o What is CBIR?

The retrieval of images from a databasebased on content features such as

colour, texture and shape.

o Reasons for its developments

Insufficiency in certain applications


32/33

Conclusion

o Applications

Finger print scanning systems

Automatic face recognition systems

Medical image databases

Trademark image registration


33/33

Conclusion

o Our Approach

Colour

Texture

Shape

o Where we are

In the phase of understanding andimplementing shape.

Presentation on cbir

Documents