Detection of compound structures using multiple hierarchical segmentations

Date
2012
Advisor
Supervisor
Co-Advisor
Co-Supervisor
Instructor
Source Title
2012 IEEE International Geoscience and Remote Sensing Symposium
Print ISSN
2153-6996
Electronic ISSN
2153-7003
Publisher
IEEE
Volume
Issue
Pages
6833 - 6836
Language
English
Type
Conference Paper
Journal Title
Journal ISSN
Volume Title
Series
Abstract

In this paper, our aim is to discover compound structures comprised of regions obtained from hierarchical segmentations of multiple spectral bands. A region adjacency graph is constructed by representing regions as vertices and connecting these vertices that are spatially close by edges. Then, dissimilarities between neighboring vertices are computed using statistical and structural features, and are assigned as edge weights. Finally, the compound structures are detected by extracting the connected components of the graph whose edges with relatively large weights are removed. Experiments using WorldView-2 images show that grouping of these vertices according to different criteria can extract high-level compound structures that cannot be obtained using traditional techniques. © 2012 IEEE.

Course
Other identifiers
Book Title
Keywords
Object detection, Compound structures, Connected component, Edge weights, Graph-based representations, Hier-archical clustering, Hierarchical segmentation, Object Detection, Region adjacency graphs, Spectral band, Structural feature, Traditional techniques, Geology, Remote sensing
Citation
Published Version (Please cite this version)