Comparison between Content –Based Image Retrieval Based on Electromagnetism-Like Mechanism” on Hindawi Publishing Corporation, Mathematical Problems in Engineering Volume 2013 available on http://dx.doi.org/10.1155/2013/782519 and
Content Based Image Retrieval using Exact Legendre Moments and Support Vector Machine on The International Journal of Multimedia and Its Application Volume 2, No.2, May 2010.
3. Articles
1. Content –Based Image Retrieval Based on
Electromagnetism-Like Mechanism” on Hindawi
Publishing Corporation, Mathematical Problems in
Engineering Volume 2013 available on
http://dx.doi.org/10.1155/2013/782519
2. Content Based Image Retrieval using Exact Legendre
Moments and Support Vector Machine on The
International Journal of Multimedia and Its
Application Volume 2, No.2, May 2010.
4. Introduction
Attention in the potential of digital images has increased
enormously over the last few years, fuelled at least in part
by the rapid growth of imaging on the World-Wide Web.
Many of collections are the product of digitizing exiting
content of analog photographs, paints, diagrams and etc…
Content-Based Image Retrieval (CBIR) is a new active
research discipline focused on computational strategies to
search for relevant images based on visual content analysis.
5. Definitions
Content Based Image Retrieval (CBIR):- is a
prominent area in image processing due to its diverse
applications in internet, multimedia, medical image
archives, and crime prevention.
Image: - Images have rich content, and each image is
described by its own features.
Image Moment: - is certain particular weighted average
(moment) of the image pixels’ intensities, or a function of
such moments, usually chosen to have some attractive
property or interpretation. (Wikipedia)
6. Articles Summary
CBIR Based on Electromagnetism-Like Mechanism
Is based on based on electromagnetism optimization technique
that follows the collective attraction-repulsion mechanism by
considering each image as electrical charge
CBIR Based on Exact Legendre Moments and Support
Vector Machine
Id based on shape using invariant image moments, viz, Moment
Invariants (MI) and Zernike Moments (ZM) are available in the
article.
7.
8. Image Representation
Electromagnetism-Like Mechanisms originally came
from the electromagnetism theory in physics, which
simulates the electromagnetism theory by considering
each particle to be an electrical charge
Exact Legendre Moments and Support Vector Machine
represent images using image moments like (Moment
Invariants, Zernike Moments, Exact Legendre
Moments and Support Vector Machine)
9. Similarity Measurement
Retrieved images are ranked by their Euclidean
distance in Electromagnetism-Like Mechanisms
Similarities are viewed on the image database as two
sets of vectors in an “n” dimensional space and
constructs a separating hyper plane that maximizes
the margin between the images relevant to query and
images not relevant to the query in Exact Legendre
and Support Vector Machine
10. Learning Mechanism
Learning is not provided in Electromagnetism-Like
Mechanism
Learning is provided in Exact Legendre Moments and
Support Vector Machine and mostly training of the
system is very important in Exact Legendre and
Support Vector Machine
11. Experiments
Electromagnetism-Like Mechanisms, experiments
were carried out on a subset of images from the Corel
Photo Gallery. This subset contains about 20,000
images of very diverse subject matter.
Exact Legendre Moments and Support Vector Machine
Corel shape database, COIL-20, 20 classes of images
with each class consisting of 72 different orientations
resulting in a total of 1440 images. And a result of
12. Results
Electromagnetism-Like Mechanisms’ algorithm is able
to advance the overall average precision to 83.57% and
64.10% using noisy or blurred query image.
Exact Legendre Moments and Support Vector Machine
algorithm is capable of 62.75% average precision.
13. Analysis
Extraction of visual features, color, texture, and shape
is an important component of CBIR plus speed and
accuracy
Exact Legendre Moments and Support Vector Machine
presents image as geometric object with approximation.
Electromagnetism-Like Mechanisms represent as
electromagnet charge.
Electromagnetism-Like Mechanisms has high precession
than Exact Legendre Moments and Support Vector
Machine
14. Conclusion
Moment invariants have been widely applied to image
pattern recognition due to its invariant features on
image translation, scaling and rotation.
Electromagnetism has both been successfully applied
to constraint project scheduling problems, vehicle
routing, array pattern optimization, image processing,
neural network training and control system.