C. Sombattheera et al. (Eds.): MIWAI 2012, LNCS 7694, pp. 176–187, 2012.
© Springer-Verlag Berlin Heidelberg 2012
Efficient Handling of 2D Image Queries Using VPC
+
-tree
Tran Cong Doi, Quan Thanh Tho, and Duong Tuan Anh
Ho Chi Minh City University of Technology, 268 Ly Thuong Kiet, District 10,
Ho Chi Minh City, Vietnam
congdoivc@gmail.com,
{qttho,dtanh}@cse.hcmut.edu.vn
Abstract. Handling queries over images is an interesting issue emerging
recently in information systems. One of the most challenging problems on that
work is how to process the image rotation efficiently since the query image and
the ones stored in the database were typically not taken from the same angles.
In this paper, an approach that employs time series representation of images is
introduced. Subsequently, Fourier Transform technique can be performed to
achieve the invariant rotation between images. Moreover, the data can be
compressed efficiently on that representation when working on huge amount of
data.
The major contribution on this work is the proposal of VPC
+
-tree, extended
from VPC-tree, a well-known structure supporting indexing and retrieving
compressed objects. The VPC
+
-tree not only supports faster and more accurate
retrieval, but it also achieves the almost ideal ratio of disc access. It is a
remarkable contribution in the field of time series data processing.
Keywords: shape matching, 2D image query, time series, invariant rotation,
Fourier transform, VPC
+
-tree.
1 Introduction
Image retrieval has been widely applied in many domains like anthropology (Brien
and Lyman, 2003), biology or medicine (Li Wei et. al., 2006). One of the most useful
applications is that doctors make observation based on the shapes of certain cells to
identify symptoms of diseases by comparing the species with the stored samples.
Thus, finding similar shapes is among the most demanded tasks on this field.
In practical situations, the shape-matching techniques employed should be
invariant with the image rotation. Besides, those approaches must deal with other
issues like the problem of huge multidimensional space used for image representation,
as well as the huge image databases need to be stored on disk for a query meanwhile
the search execution should be done in a quick period.
Basically, typical image processing techniques can deal with the problem of image
rotation (Adamek and Connor, 2004). However, those techniques suffer from the high
computational costs and therefore they are hardly applied in a real-time application. In
recent years, there is a remarkable emerging technique of using time series for 2D
image representation and processing. The representative time series information can
be further processed using basic technique like Fourier transform to make it invariant