Content-Based Retrieval of Aurora Images Based on the Hierarchical Representation - Advanced Concepts for Intelligent Vision Systems - page 248

Content-Based Retrieval of Aurora Images
Based on the Hierarchical Representation
Soo K. Kim 1 and Heggere S. Ranganath 2
1 Clarion University of Pennsylvania
Clarion, PA 16214, USA
[e-mail]
2 University of Alabama in Huntsville
Huntsville, AL 35899, USA
[e-mail]
Abstract. The boundary based image segmentation and representation system
takes a thinned edge image and produces a unique hierarchical representation
using a graph/tree data structure. The feature extraction algorithms have been
developed to obtain geometric features by directly processing the graph/tree
hierarchical data structure for diverse image processing and interpretation ap-
plications. This paper describes a content-based image retrieval system for the
retrieval of aurora images utilizing the graph/tree hierarchical representation
and the associated geometric feature extraction algorithms which extract fea-
tures for the purpose of classification. The experimental results which prove
that the hierarchical representation supports the fast and reliable computation of
several geometric features which are useful for content based image retrieval
are presented.
1 Introduction
The content based retrieval of aurora images is a subject of great interest to space
scientists. Aurora is a significant phenomenon in the polar region of the Earth [1, 2].
It is a result of interaction between the solar wind and the Earth's magnetic field.
Auroral events are monitored on the global scale at the Far Ultraviolet (FUV) spec-
trum by the Ultraviolet Imager (UVI) on board the POLAR satellite. Detecting an
oval boundary of aurora is not a trivial problem because the distinction between auro-
ra and background is not clear in most cases. In addition, the existence of dayglow
emission significantly limits the automatic determination of the location and the size
of auroral ovals. Auroral morphological parameters include the location and shape of
the boundaries, the size of auroral ovals, and the evolution of intensified aurora arc
regions during substorm events. The shape of aurora is dynamic and changes depend-
ing on the factors such as the date, time, the satellite position, etc.
Three specific types of aurora images that are of great interest to scientists have
been identified. The first type, Type1 , is defined as the aurora that has the very high
magnetic latitude activity, called transpolar arc, close to the pole [1]. An example of
the Type1 aurora is shown in Fig. 1 (a). The second type, Type2 , is defined as the
aurora that is thick. In this aurora, a strong magnetic storm, often referred to as a sub-
storm, might be present. A substorm starts as a bright spot on the auroral oval. The