Multidimensional Indexing Structures for Content‐Based Retrieval

Castelli, Vittorio

doi:10.1002/0471224634.ch14

Cited by 12 publications

(19 citation statements)

References 108 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We have studied the effect of CSVD in processing k-NN queries on disk-resident high-dimensional indexing methods: R*-trees, SR-trees, and hybrid trees, as well (see [2] for a description). SR-trees outperform the other two index structures when the number of page accesses per query is used as a metric [18].…”

Section: Discussionmentioning

confidence: 99%

“…Similarity of two objects is determined by the proximity of the endpoints of the N -dimensional feature vectors representing them. Proximity is determined by the Euclidean distance or some other similarity measures [2].…”

Section: Introductionmentioning

confidence: 99%

“…CPU time is dominated by the cost of computing Euclidean distances, since the cost of inserting nearest neighbor candidates into a heap is negligibly small. It is especially important to minimize the number of objects inspected for more complex similarity measures [2].…”

Section: Introductionmentioning

confidence: 99%

“…The processing of k-NN queries with a reduced number of dimensions is less costly, but yields approximate results, as quantified by recall and precision [5] (see Section 2). Most multidimensional indexing structures [6] lose their efficiency when processing k-NN queries on high dimensional data [5,2], so dimensionality reduction is another method to cope with the "curse of dimensionality". This paper is concerned with the cost of reducing CPU time using linear scans of associated datasets.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Exact k-NN queries on clustered SVD datasets

Thomasian

Zhang

2005

Information Processing Letters

View full text Add to dashboard Cite

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Exact k-NN queries on clustered SVD datasets

Thomasian

Zhang

2005

Information Processing Letters

View full text Add to dashboard Cite

“…The similarity measure depends on user's criteria and on the representation of image features (Li et al, 2002). In CBIR systems, the main method consists in representing each image of the database by a multidimensional feature vector, also called descriptor (Castelli, 2002). The similarity between images is computed from the distance between their feature vectors (Di Gesú et al, 1999;Rubner et al, 2001) such that proximity in the multi-dimensional space reflects the similarity of the images.…”

Section: Introductionmentioning

confidence: 99%

Δ-distance: A family of dissimilarity metrics between images represented by multi-level feature vectors

2006

View full text Add to dashboard Cite

This article presents the -distance, a family of distances between images recursively decomposed into segments and represented by multi-level feature vectors. Such a structure is a quad, a quin or a nona-tree resulting from a fixed and arbitrary image partition or from an image segmentation process. It handles positional information of image features (e.g. color, texture or shape). -distance is the generalized form of dissimilarity measures between multi-level feature vectors. Using different weights on tree nodes and different distances between nodes, distances between trees or visual similarity between images can be computed based on the general definition of . In this article, we present three -based distance families: two families of distances between tree structures, called T -distance (T for Tree) and S-distance (S for Segment), and a family of visual distances between images, called V-distance (V for Visual). The V-distance visually compares two images using their tree representation and the other two distances compare the tree structures resulting from image segmentation. Moreover, we show how existing distances between multi-level feature vectors appear to be particular cases of the -distance.Keywords Image database . Distance between quad/quin or nona-trees . Similarity of images . Similarity of image segments . Content-based image retrieval

show abstract

Towards the semantic extraction of digital signatures for librarian image‐identification purposes

Poulos

Bokos

Vaioulis

2008

J. Am. Soc. Inf. Sci.

View full text Add to dashboard Cite

In study we attempt to solve the problem of determining how authentic published images on the Internet are, and to what degree they may be identified by comparison to the original image. The technique proposed aims to serve the new requirements of libraries. One of these is the development of computational tools for the control and preservation of intellectual property such as digital objects, and especially of digital images. For this purpose, this article proposes the use of a serial number extracted using a previously tested semantic-properties method. This method, based on the multilayers of a set of arithmetic points, assures the following two properties: the uniqueness of the final extracted number, and the semantic dependence of this number on the image used as the method's input. The major advantage of this method is that it can serve as the authentication for a published image or detect partial modifications to a reliable degree. Also, it requires the better of the known hash functions that the digital-signature schemes use, and produces alphanumeric strings for checking authenticity and the degree of similarity between an unknown image and an original image. As an example of a possible application, this article suggests that this method could be incorporated into the well-known DOI system in order to provide a reliable tool for identification and comparison of digital images.

show abstract

Multidimensional Indexing Structures for Content‐Based Retrieval

Cited by 12 publications

References 108 publications

Exact k-NN queries on clustered SVD datasets

Exact k-NN queries on clustered SVD datasets

Δ-distance: A family of dissimilarity metrics between images represented by multi-level feature vectors

Towards the semantic extraction of digital signatures for librarian image‐identification purposes

Contact Info

Product

Resources

About