Professional Documents
Culture Documents
CONTENT-BASED
IMAGE RETRIEVAL
SYSTEM
USING SKETCHES
GROUP IV
GEETHU P T
HAFSA HASSAN
HONEY MERRIN SAM
SHIBIJA K
TABLE OF
CONTENTS
INTRODUCTION
CURRENT SITUATION
PROPOSED SOLUTION
ALGORITHM USED
PURPOSE
GLOBAL STRUCTURE AND
SUBSYSTEM
FUNCTIONAL REQUIREMENTS
PERFORMANCE MATRICES
FEATURE EXTRACTION
USE CASE DIAGRAM
DATA FLOW DIAGRAM
ADVANTAGES
SBIR VS TEXT-BASED IMAGE
RETRIEVAL
RESEARCH CHALLENGES
APPLICATIONS
FUTURE GROWTH
CONCLUSION
INTRODUCTION
The sketch based image retrieval is one
of most popular, rising research areas of
the digital image processing.
Goal of SBIR is to extract visual content
like colour , text, or shape. Introduces
design based on a free hand sketch
Making search more efficient hereby
Test results show that the sketch based
system allows users an intuitive access
to search-tools.
CURRENT
SITUATION
In earlier days,
image retrieval from large
image database can be done by following
ways.
Automatic
image
annotation
and
retrieval using cross media relevance
models
Concept based query expansion
Query system bridging the semantic gap
for large image databases
Ontology-based query expansion widget
for information retrieval
Detecting image purpose in world-wide
web documents
PROPOSED
SOLUTION
Relevance feedback is an
interactive
process that starts with normal CBIR.
Input
Query
Featur
e
Extract
ion
Simila
rity
Measu
re
Query
updat
e
Users
Feedba
ck
Retrie
val
Result
Find all
Images
?
Final
Retriev
al
Result
ALGORITHM
USED
In this system an efficient image retrieval
algorithm based on CCM (Colour Cooccurrence Matrix) is proposed.
The CCM for each pixel of an image is
found using the Hue Saturation Value
(HSV) of the pixel and then compared with
CCM of the images in the database and the
images are retrieved
Hue
Query
Image
Extract
HSV
Formula
te CCM
Compar
e and
Match
Query
Image
Extract
HSV
Formula
te CCM
Retrieve
d Image
PURPOSE
The goal of this paper is to develop a
SBIR search engine, which with free
hand sketch content can be retrieved.
The most important task is to bridge
the gap between the free hand sketch
and the picture.
Introducing this system into search
engines makes corporate world and
even other users bit more efficient in
retrieval of data effectively.
Preprocessed
image
image
Displaying
Subsystem
Preprocessing
Subsystem
Result
Feature Vector
generating subsystem
Feature vector
Retrieval
subsystem
Stock index
Database Management
subsystem
Input
Image
Preprocessing
subsystem
Output
Processed
Image
Thesystemisfordatabasescontainingsimple
images
FUNCTIONAL
REQUIREMENTS
1.LOGIN MODULE
2. ADMIN MODULE
3. USERS MODULE
PERFORMANCE
is
essentially
an
information
METRICES
SBIR
retrieval problem. Two of the most
popular evaluation measures are the ,
PRECISION
The
precision
measures
the
proportion of the total images
retrieved which are relevant
to the query.
precision = number of relevant
images retrieved
Total
RECALL
The recall measure is defined as the
fraction of the all relevant images.
Recall =
images
FEATURE
1. COLOUREXTRACTION
Colour is the most extensively used
visual content for image retrieval.
The most common primary colours in
computing are red, green and blue
(e.g. colours used in a monitor).
Usually colours are defined in three
dimensional colour spaces.
In image retrieval systems, colour
histogram is the most commonly
used feature representation.
The colour histogram describes the
proportion of pixels of each colour in
an
image
with
simple
and
2. TEXTURE
Texture refers to the visual patterns
with properties of homogeneity that
do not result from the presence of a
single colour or intensity.
It is that innate property of all
surfaces that
describes visual
patterns such as; clouds, leaves,
bricks, fabric, etc.
It contains information about the
structural arrangement of surfaces
and
their
relationship
to
the
surrounding environment.
Texture
properties
include
coarseness, contrast, directionality,
regularity and roughness
3. SHAPE
Shape is an important criterion for
matching objects based on their
profile and physical structure.
Shape does not refer to the shape of
an image but to the shape of a
particular region that is being sought
out.
Shape features can represent spatial
information that is not represented
by colour or texture.
It contains all
the geometrical
information of an object in the image
which does not change generally
change even when orientation or
RESULTS OF FEATURE
EXTRACTION:
USE CASE
DIAGRAM
System
Index data base images
System
Upload/search
images
Sketch images
Data
base
Retrieve
images
User
Query update
DATA FLOW
DIAGRAM
Indexing
and
searchin
g
User
Calcul
ate
result
Display
Image
Data
base
Feature
Extraction
Image
Datab
ase
Compute
similarity
Measure
Visualizati
on
ADVANTAGES
Make convenient to retrieve data or
images based on sketches so that even
illiterates , who do not know to write
text can also make use of system
effectively.
Introducing this system into search
engines makes corporate world and even
other users bit more efficient in retrieval
of data effectively.
Visual features, such as colour, texture,
and shape information, of images are
extracted automatically
SBIR VS TEXT-BASED
IMAGE RETRIVEL
SYSTEM
Image retrieval
algorithms roughly belong
to two categories:
1. Text-based approaches
The
text-based
approaches
associate keywords with each
stored image.
2. Content-based methods
Here retrieval of images is
guided by providing a query
image or a sketch generated by a
HOW TO DESCRIBE
THIS IMAGE ?
WHAT IS THIS
IMAGE?
SKY
RIVE
R
COCONUT
TREES
Gira
ffe
Text based
image
retrieval
system
Image
datab
ase
CBIR
System
Image
datab
ase
RESEARCH
The implementation
of SBIR systems raises
CHALLANGES
several research challenges. Some of these
are:
Understanding image users needs and
information-seeking behaviour.
Investigating new user interfaces for
annotating, browsing, and searching
based on image content .
New
tools
for
marking/annotating
images (and their regions)
Matching query and stored images in a
way that reflects human similarity
judgments
APPLICATIONS
1. MEDICAL APPLICATIONS
Three large domains can instantly
take advantage of SBIR techniques:
Teaching
Research
Diagnostics
2. BIODIVERSITY INFORMATION
SYSTEMS
Ideally,
Biodiversity
Information
Systems
(BIS)
should
help
researchers to enhance or complete
their knowledge and understanding
about species and their habitats by
3.
DIGITAL LIBRARIES
One example is the digital museum
of butterflies aimed at building a
digital
collection
of
Taiwanese
butterflies.
This digital library includes a module
responsible for content-based image
retrieval based on colour, texture,
and patterns.
FUTURE GROWTH
1. WEB ORIENTED
To better organize and retrieve the
almost unlimited information, web
based search engines are highly
desired.
Such solutions exists for text
based information's
2. HIGH DIMENSIONAL INDEXING
Most currently existing research
prototype systems only handle
hundreds or at most thousands of
images.
CONCLUSION
The area of SBIR is a hybrid research area
that
requires
knowledge
of
both
computer vision and of database systems.
The technology is exciting but immature.
Among the objectives of this paper ,two
main aspects were taken into account.
1. The retrieval process has to be highly
interactive.
2. The robustness of the method.
Frequently
updated
shared
image
database and the regular comparison of
system performances would be of great