31-10-2016, 02:45 PM
CONTENT BASED IMAGE RETRIEVALTECHNIQUE BASED ON TEXTURE AND SHAPE
ANALYSIS USING WAVELET FEATURE AND CLUSTERING MODEL
1462520477-FINALFINAL.pdf (Size: 1.16 MB / Downloads: 6)
ABSTRACT
The project proposes the image retrieval technique based on GLCM and
Shape features. The main target of CBIR is to get accurate results with
lower computational time. The need for efficient content-based image
retrieval has increased tremendously in many application areas such as
biomedicine, military, commerce, education, and web image classification
and searching. Content-based Image Retrieval (CBIR) technology
overcomes the defects of traditional text-based image retrieval technology,
such as heavy workload and strong subjectivity. It makes full use
of image content features which are analyzed and extracted automatically
by computer to achieve the effective retrieval Using a single feature
for image retrieval cannot be a good solution for the accuracy and effi-
ciency. This paper discusses on the comparative method used in colour
histogram based on two major methods used frequently in CBIR which
are; normal colour histogram using GLCM, and colour histogram using
K-Means. Using Euclidean distance, similarity between queried image
and the candidate images are calculated. The colour histogram with KMeans
method had high accuracy and precise compared to GLCM. Future
work will be made to add more features that are famous in CBIR
which is texture features extraction using discrete wavelet based entropy
measurement in order to get better results.
INTRODUCTION
Image retrieval techniques are useful in many image-processing applications.
Content-based image retrieval systems work with whole images and searching is based
on comparison of the query. General techniques for image retrieval are color, texture
and shape. These techniques are applied to get an image from the image database. They
are not concerned with the various resolutions of the images, size and spatial color
distribution. Hence all these methods are not appropriate to the art image retrieval.
Moreover shape based retrievals are useful only in the limited domain. The content and
metadata based system gives images using an effective image retrieval technique. Many
other image retrieval systems use global features like color, shape and texture. But the
prior results say there are too many false positives while using those global features to
search for similar images. Hence we give the new view of image retrieval system using
both content and metadata.
1.1 BACKGROUND
1.1.1 The Growth OF Digital Imaging
The use of images in human communication is hardly new our cave-dwelling
ancestors painted pictures on the walls of their caves, and the use of maps and building
plans to convey information almost certainly dates back to pre-Roman times. But the
twentieth century has witnessed unparalleled growth in the number, availability and importance
of images in all walks of life. Images now play a crucial role in fields as diverse
as medicine, journalism, advertising, design, education and entertainment. Technology,
in the form of inventions such as photography and television, has played a major role
in facilitating the capture and communication of image data. But the real engine of the
imaging revolution has been the computer, bringing with it a range of techniques for
digital image capture, processing, storage and transmission which would surely have
startled even pioneers like John Logie Baird. The involvement of computers in imaging
can be dated back to 1965, with Ivan Sutherland’s Sketchpad project, which demonstrated
the feasibility of computerized creation, manipulation and storage of images,
though the high cost of hardware limited their use until the mid-1980s. Once computerized
imaging became affordable (thanks largely to the development of a mass market
for computer games), it soon penetrated into areas traditionally depending heavily on
images for communication, such as engineering, architecture and medicine. Photograph
libraries, art galleries and museums, too, began to see the advantages of making their
collections available in electronic form. The creation of the World-Wide Web in the
early 1990s, enabling users to access data in a variety of media from anywhere on the
planet, has provided a further massive stimulus to the exploitation of digital images.
The number of images available on the Web was recently estimated to be between 10
and 30 million [Sclaroff et al, 1997] a figure in which some observers consider to be a
significant underestimate.
1.1.2 The Need For Image Data Management
The process of digitization does not in itself make image collections easier
to manage. Some form of cataloguing and indexing is still necessary as the only difference
being that much of the required information can now potentially be derived
automatically from the images themselves. The extent to which this potential is currently
being realized is discussed below. The need for efficient storage and retrieval
of images recognized by managers of large image collections such as picture libraries
and design archives for many years was reinforced by a workshop sponsored by the
USA’s National Science Foundation in 1992 [Jain, 1993]. After examining the issues
involved in managing visual information in some depth, the participants concluded that
images were indeed likely to play an increasingly important role in electronically mediated
communication. However, significant research advances, involving collaboration
between a numbers of disciplines, would be needed before image providers could take
full advantage of the opportunities offered. They identified a number of critical areas
where research was needed, including data representation, feature extractions and indexing,
image query matching and user interfacing. One of the main problems they
highlighted was the difficulty of locating a desired image in a large and varied collec-
2
tion. While it is perfectly feasible to identify a desired image from a small collection
simply by browsing, more effective techniques are needed with collections containing
thousands of items. Journalists requesting photographs of a particular type of event,
designers looking for materials with a particular color or texture, and engineers looking
for drawings of a particular type of part, all need some form of access by image
content. The existence and continuing use of detailed classification schemes such as
ICONCLASS [Gordon, 1990] for art images, and the Opitz code [Opitz et al, 1969] for
machined parts, reinforces this message.
1.2 PROBLEM STATEMENT
The goals for this thesis have been the following.The primary goal our project
is to reduce the computation time and user interaction. The conventional Content Based
Image Retrieval (CBIR) systems also display the large amount of results at the end of
the process this will drove the user to spend more time to analyze the output images.
In our proposed system we compute texture feature and color feature for compute the
similarity between query and database images. This integrated approach will reduce the
output results to a certain levels based on the user threshold value. The secondary goal is
to reduce semantic gap between high level concepts and low level features. Generally
the content based image retrieval systems compute the similarity between the query
image and the database images. Hence there might be chances for unexpected results at
the end the retrieval process. The novel clustering technique cluster the output images
and select one representative image from each clusters. A third goal is to evaluate their
performance with regard to speed and accuracy. These properties were chosen because
they have the greatest impact on the implementation effort. A final goal has been to
design and implement an algorithm. This should be done in high-level language or
Matlab. The source code should be easy to understand so that it can serve as a reference
on the standard for designers that need to implement real-time motion detection.
3
1.3 LITERATURE SURVEY
Title 1: Content- Content-Based Image Retrieval using Color Moment and Gabor
Based Image Retrieval using Color Moment and Gabor Texture Feature.
Author:S. Mangijao Singh , K. Hemachandran.
Year: 1 September 2014.
Abstract: Content based image retrieval (CBIR) has become one of the most active
research areas in the past few years. Many indexing techniques are based on global
feature distributions. However, these global distributions have limited discriminating
power because they are unable to capture local image information. In this paper, we
propose a content-based image retrieval method which combines color and texture features.
To improve the discriminating power of color indexing, we encode a minimal
amount of spatial information in the color index. As its color features, an image is divided
horizontally into three equal non-overlapping regions. From each region in the
image, we extract the first three moments of the color distribution, from each color
channel and store them in the index i.e., for a HSV color space, we store 27 floating
point numbers per image. For texture feature, Gabor texture descriptors are adopted.
We assign weights to each feature respectively and calculate the similarity with combined
features of color and texture using Canberra distance as similarity measure.
Title 2: Content Based Image Retrieval Using Color Histogram.
Author:A.Ramesh Kumar, D.Saravanan.
Year: 15 December 2013.
Abstract: Content-based image retrieval (CBIR) scheme searches the most similar images
of a query image that involves in comparing the feature vectors of all the images
in the database with that of the query image using some pre-selected similarity measure,
and then sorting of the results. On querying an image, a reduced set of candidate
images which have the same Grid Code as that of the query image is obtained. The
color histogram for an image is constructed by quantizing the colors within the image
and counting the number of pixels of each color. The feature vector of an image can be
derived from the histograms of its color components.
4
Title 3: Content-Based Image Retrieval Using Wavelet Packets and Fuzzy Spatial Relations.
Author: Minakshi Banerjee and Malay K. Kundu.
Year: 13 December 2006.
Abstract: This paper proposes a region based approach for image re- trieval. We develop
an algorithm to segment an image into fuzzy regions based on coefficients of multiscale
wavelet packet transform. The wavelet based features are clustered using fuzzy
C-means algorithm. The final cluster centroids which are the representative points, signify
the color and texture properties of the preassigned number of classes. Fuzzy Topological
relationships are computed from the final fuzzy partition matrix. The color and
texture properties as indicated by centroids and spatial relations between the segmented
regions are used together to provide overall characterization of an image. The closeness
between two images are estimated from these properties. The performance of the
system is demonstrated using different set of examples from general purpose image
database to prove that, our algorithm can be used to generate meaningful descriptions
about the contents of the images.
Title 4: A Novel Active Learning Method in Relevance Feedback for Content-Based
Image Retrieval.
Author: Begum Demir,Lorenzo Bruzzone.
Year: 8th January 2015.
Abstract: Conventional relevance feedback (RF) schemes improve the performance
of content-based image retrieval (CBIR) requiring the user to annotate a large number
of images. To reduce the labeling effort of the user, this paper presents a novel active
learning (AL) method to drive RF for retrieving remote sensing images from large
archives in the framework of the support vector machine classifier. The proposed AL
method is specifically designed for CBIR and defines an effective and as small as possible
set of relevant and irrelevant images with regard to a general query image by jointly
evaluating three segments of uncertainity, diversity and density.
Content Based Image Retrieval
In contrast to the text-based approach of the systems described in section
above, CBIR operates on a totally different principle, retrieving stored images from a
collection by comparing features automatically extracted from the images themselves.
The commonest features used are mathematical measures of color, texture or shape;
hence virtually all-current CBIR systems, whether commercial or experimental, operate
at level 1. A typical system that allows users to formulate queries by submitting an
example of the type of image being sought, though some offer alternatives such as
selection from a palette or sketch input. The system then identifies those stored images
whose feature values match those of the query most closely, and displays thumbnails of
these images on the screen. Some of the more commonly used types of feature used for
image retrieval are described below.
2.1.1 Color Feature Based Retrieval
Several methods for retrieving images on the basis of color similarity have
been described in the literature, but most are variations on the same basic idea. Each
image added to the collection is analyzed to compute a color histogram, which shows
the proportion of pixels of each color within the image. The color histogram for each
image is then stored in the database. At search time, the user can either specify the
desired proportion of each color (75 percent of olive green and 25 percent of red, for
example), or submit an example image from which a color histogram is calculated.
Either way, the matching process then retrieves those images whose color histograms
match those of the query most closely. The matching technique most commonly used,
histogram intersection, was first developed by Swain and Ballard . Variants of this
technique are now used in a high proportion of current CBIR systems. Methods of
improving on Swain and Ballard’s original technique include the use of cumulative color histograms and combining histogram intersection with some element of spatial
matching , and the use of region-based color querying The results from some of these
systems can look quite impressive.
a)RGB Color model
The RGB Coordinates system and color model were represented in Figure 2.2 and
2.3 respectively. This system defines the color model that is used in most color CRT
monitors and color raster graphics. They are considered the "additive primaries" since
the colors are added together to produce the desired color. The RGB model uses the
cartesian coordinate system.Notice the diagonal from (0,0,0) black to (1,1,1) white
which represents the grey-scale.
b)HSV Color model
The HSV stands for the Hue, Saturation, and Value based on the artists (Tint, Shade,
and Tone). The coordinate system in a hexacone in Figure 2.4 and Figure 2.5 a view
of the HSV color model. The Value represents intensity of a color, which is decoupled
from the color information in the represented image. The hue and saturation components
are intimately related to the way human eye perceives color resulting in image
processing algorithms with physiological basis.
As hue varies from 0 to 1.0, the corresponding colors vary from red, through yellow,
green, cyan, blue, and magenta, back to red, so that there are actually red values both
at 0 and 1.0. As saturation varies from 0 to 1.0, the corresponding colors (hues) vary
from unsaturated (shades of gray) to fully saturated (no white component). As value, or
brightness, varies from 0 to 1.0, the corresponding colors become increasingly brighter.
c)Color conversion
In order to use a good color space for a specific application, color conversion is
needed between color spaces. The good color space for image retrieval system should
preserve the perceived color differences. In other words, the numerical Euclidean difference
should approximate the human perceived difference.
Histogram-Based Image Search
The color histogram for an image is constructed by counting the number of
pixels of each color. Retrieval from image databases using color histograms has been
investigated in [tools, fully, automated]. In these studies the developments of the extraction
algorithms follow a similar progression: (1) selection of a color space, (2) quantization
of the color space, (3) computation of histograms, (4) derivation of the histogram
distance function, (5) identification of indexing shortcuts. Each of these steps may be
crucial towards developing a successful algorithm.
There are several difficulties with histogram based retrieval. The first of these is
the high dimensionality of the color histograms. Even with drastic quantization of the
color space, the image histogram feature spaces can occupy over 100 dimensions in
real valued space. This high dimensionality ensures that methods of feature reduction,
pre-filtering and hierarchical indexing must be implemented. The large dimensionality
also increases the complexity and computation of the distance function. It particularly
complicates ‘cross’ distance functions that include the perceptual distance between histogram bins .
Color Histogram
An image histogram refers to the probability mass function of the image intensities.
This is extended for color images to capture the joint probabilities of the intensities of
the three color channels. More formally, the color histogram is represented in,eq(3)
hA,B,C(a, b, c) = N.P rob(A = a, B = b, C = c) ..(3)
Where A , B and C represent the three color channels (R,G,B or H,S,V) and N is
the number of pixels in the image. Computationally, the color histogram is formed
by discretizing the colors within an image and counting the number of pixels of each
color. Since the typical computer represents color images with up to 224 colors, this
process generally requires substantial quantization of the color space. The main issues
regarding the use of color histograms for indexing involve the choice of color space
and quantization of the color space. When a perceptually uniform color space is chosen
uniform quantization may be appropriate. If a non-uniform color space is chosen, then
non-uniform quantization may be needed. Often practical considerations, such as to be
compatible with the workstation display, encourage the selections of uniform quantization
and RGB color space. The color histogram can be thought of as a set of vectors. For
11
gray-scale images these are two dimensional vectors. One dimension gives the value
of the gray-level and the other the count of pixels at the gray-level. For color images
the color histograms are composed of 4-D vectors. This makes color histograms very
difficult to visualize. There are several lossy approaches for viewing color histograms,
one of the easiest is to view separately the histograms of the color channels. This type
of visualization does illustrate some of the salient features of the color histogram .
2.1.4 Color Uniformity
The RGB color space is far from being perceptually uniform. To obtain a good
color representation of the image by uniformly sampling the RGB space it is necessary
to select the quantization step sizes to be fine enough such that distinct colors are not
assigned to the same bin. The drawback is that oversampling at the same time produces
a larger set of colors than may be needed. The increase in the number of bins in the
histogram impacts performance of database retrieval. Large sized histograms become
computationally unwieldy, especially when distance functions are computed for many
items in the database. Furthermore, as we shall see in the next section, to have finer but
not perceptually uniform sampling of colors negatively impacts retrieval effectiveness.
However, the HSV color space mentioned earlier offers improved perceptual uniformity.
It represents with equal emphasis the three color variants that characterize color:
Hue, Saturation and Value (Intensity). This separation is attractive because color image
processing performed independently on the color channels does not introduce false colors.
Furthermore, it is easier to compensate for many artifacts and color distortions. For
example, lighting and shading artifacts are typically be isolated to the lightness channel.
But this color space is often inconvenient due to the non-linearity in forward and
reverse transformation with RGB space.
2.1.5 Color Histogram Discrimination
There are several distance formulas for measuring the similarity of color histograms.
In general, the techniques for comparing probability distributions, such as
the kolmogoroff-smirnov test are not appropriate for color histograms. This is because
visual perception determines similarity rather than closeness of the probability distribu-
12
tions. Essentially, the color distance formulas arrive at a measure of similarity between
images based on the perception of color content. Three distance formulas that have been
used for image retrieval including histogram euclidean distance, histogram intersection
and histogram quadratic (cross) distance