25-07-2012, 02:48 PM
CONTENT BASED IMAGE RETRIEVAL THROUGH OBJECT EXTRACTION
CONTENT BASED IMAGE RETRIEVAL.pdf (Size: 283.34 KB / Downloads: 51)
Abstract-
We propose a content based image retrieval system based on object extraction through image
segmentation. A general and powerful multiscale segmentation algorithm automates the segmentation process,
the output of which is assigned novel colour and texture descriptors which are both efficient and effective. Query
strategies consisting of a semi-automated and a fully automated mode are developed which are shown to produce
good results. We then show the superiority of our approach over the global histogram approach which proves that
the ability to access images at the level of objects is essential for CBIR.
Introduction
Image retrieval has traditionally been based on
manual caption insertion describing the scene which
can then be searched using keywords. Caption
insertion is a very subjective procedure and quickly
becomes extremely tedious and time consuming,
especially for large image databases which are
becoming ever more common with the growing
availability of digital cameras and scanners. There
is thus an urgent need for effective content-based
image retrieval (CBIR) systems.
We believe the key to effective CBIR performance
lies in the ability to access the image at the level of
objects. This is because users generally want to
search for images containing particular object(s) of
interest and thus the ability to represent, index and
query images at the level of objects is critical [3].In
this paper, we present a framework for CBIR based
on unsupervised segmentation of images into
classes and querying using properties of these
classes. As these segmented classes are
homogeneous in some sense (in our case, colour
and texture), they correlate well with the identity of
objects. By decomposing images as combinations
of objects in this manner, querying becomes more
meaningful and intuitive than it is with global image
properties. This is obviously true for images with
distinct foreground objects but the rationale also
holds for ‘background’ images where no interesting
foreground objects are present. Images belonging
to the latter category can be thought of consisting of
combinations of classes with homogeneous colour
and texture (for example, images of the seaside
generally consist of the beach and the sea, images
of sunset scenes generally consist of the reddish
sky and dark silhouettes and so forth) and querying
is made more effective by being based on these
class combinations which characterise the scene.
- Decomposing an image by segmentation into
classes corresponding to ‘objects’
In our CBIR implementation, images are firstly segmented
based on joint colour and textural features
using our previously developed unsupervised
multiscale segmentation algorithm [6, 7]. The
segmentation process is completely unsupervised
and performed off-line for each image. Following
this, we represent each image using effective and
compact colour and textural descriptors of its
classes. We then structure the descriptor database
following a relational model which allows its
implementation on powerful relational database
engines. Class attribute queries are process a
Fig. 2- Typical segmentation maps of images
parallel strategy which results in significant speedContent
based image retrieval through object extraction
up in the retrieval process if parallel processor machines
are used.
In Section 2, we will briefly describe the
segmentation algorithm employed. We will then
discuss the descriptors assigned to each class in
Section 3. In Section 4, we present our query
strategy as well as preliminary results from queries
on our image database testbed consisting of various
natural images.
Unsupervised Segmentation
Our unsupervised segmentation algorithm involves
the following steps:
1. Normalised colour and texture features (three for
colour and two for texture) are mapped to a multidimensional
feature space. Spatial information is
incorporated into the process by including spatial
features into the feature space. The colour space
used is S-CIE , the spatial extension of the
perceptual uniform CIE , originally developed
by Zhang and Wandell [10]. This colour space takes
into account the appearance of fine-patterned
colours on the human visual system. Textural
features meanwhile are generated using the
logarithm of the energies of the 2-D complex
wavelet coefficients [8]1 and taking the top two
principal components.