Image Retrieval with Segmentation-Based Query.

Autor: Hutchison, David, Kanade, Takeo, Kittler, Josef, Kleinberg, Jon M., Mattern, Friedemann, Mitchell, John C., Naor, Moni, Nierstrasz, Oscar, Rangan, C. Pandu, Steffen, Bernhard, Sudan, Madhu, Terzopoulos, Demetri, Tygar, Doug, Vardi, Moshe Y., Weikum, Gerhard, Marchand-Maillet, Stéphane, Bruno, Eric, Nürnberger, Andreas, Detyniecki, Marcin, Chupikov, Andrew
Zdroj: Adaptive Multimedia Retrieval: User, Context & Feedback; 2007, p207-221, 15p
Abstrakt: Interest in digital images content has increased enormously over the last few years. Segmentation algorithms are used to extract region-based descriptions of an image and provide an input to higher level image processing, e.g. for content-based image retrieval (CBIR). Frequently it is difficult even for a user to single out representative regions or its combinations. Partitions and coverings of an image and range of gray levels (colors) are ones of principal constructive objects for an analysis. Their processing creates the necessary prerequisites to synthesize new features for CBIR and to consider redundancy and deficiency of information as well as its multiple meaning for totally correct and complete segmentation of complex scenes. The paper is dedicated to theoretical and experimental exploration of coverings and partitions produced by multithresholding segmentation. [ABSTRACT FROM AUTHOR]
Databáze: Supplemental Index