Efficient color histogram indexing for quadratic form distance functions

Hafner, James Lee; Sawhney, Harpreet; Equitz, W.; Flickner, Myron; Niblack, W.

doi:10.1109/34.391417

Cited by 682 publications

(317 citation statements)

References 7 publications

Supporting

Mentioning

307

Contrasting

Unclassified

Order By: Relevance

“…The experiments in [7] had shown that the fidelity measure tends to give sharper results if two density functions to be compared are quite close each other, comparing to χ 2 [8] and Jensen-Shannon divergence (JS) [5] measures. The quadratic form distance (QF) [4] can be the best solution because its similarity matrix may consider the closeness between the training colours.…”

Section: Texture Similarity Measurementioning

confidence: 99%

Detection of a Single Texture in Colour Images

Gimel’farb

2004

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. Detection of regions similar to a single given texture in arbitrary colour images is difficult for conventional supervised or unsupervised segmentation techniques. We introduce a novel partially supervised algorithm that solves this problem using similarity between local statistics on different levels of pyramidal representations of the texture and the image. Most characteristic statistics for the texture are estimated in accord with a generic Gibbs random field model with spatially homogeneous pairwise pixel interactions. Empirical distributions of the self-similarity values for the texture itself are used to separate the desired texture from an arbitrary background. Experiments with different images, including aerial images of the Earth's surface, show this algorithm effectively detects regions with spatially homogeneous and weakly homogeneous textures.

show abstract

Section: Texture Similarity Measurementioning

confidence: 99%

Detection of a Single Texture in Colour Images

Gimel’farb

2004

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…The first representation was a 64-dimensional color histogram. In this case, we used the weighted distance between those color histograms, represented as a quadratic form distance function as described for example in [11]. The second representation were segmentation trees.…”

Section: Clustering Images By Multiple Representationsmentioning

confidence: 99%

Clustering Multi-represented Objects with Noise

Kailing¹,

Kriegel²,

Pryakhin³

et al. 2004

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. Traditional clustering algorithms are based on one representation space, usually a vector space. However, in a variety of modern applications, multiple representations exist for each object. Molecules for example are characterized by an amino acid sequence, a secondary structure and a 3D representation. In this paper, we present an efficient density-based approach to cluster such multi-represented data, taking all available representations into account. We propose two different techniques to combine the information of all available representations dependent on the application. The evaluation part shows that our approach is superior to existing techniques.

show abstract

“…For each pair of images in group A, the distance which represents the degree of dissimilarity between the two images is calculated based on their image features. In the calculation of the distance, we do not adopt the Euclid distance but the distance which considers the proximity in the color space [8]. 3.…”

Section: Selection Partmentioning

confidence: 99%

A Fast Image-Gathering System on the World-Wide Web Using a PC Cluster

Yanai

Shindo

Noshita

2001

Web Intelligence: Research and Development

View full text Add to dashboard Cite

Abstract. Thanks to the recent explosive progress of WWW (WorldWide Web), we can easily access a large number of images from WWW. There are, however, no established methods to make use of WWW as a large image database. In this paper, we describe an automatic imagegathering system from WWW, in which we use both keywords and image features. By exploiting some existing keyword-based search engines and selecting images by their image features, our system obtains, with high accuracy, images that are strongly related to query keywords. This system has been implemented on a parallel PC cluster, which enables us to gather more than one hundred images from WWW in about one minute.

show abstract

Efficient color histogram indexing for quadratic form distance functions

Cited by 682 publications

References 7 publications

Detection of a Single Texture in Colour Images

Detection of a Single Texture in Colour Images

Clustering Multi-represented Objects with Noise

A Fast Image-Gathering System on the World-Wide Web Using a PC Cluster

Contact Info

Product

Resources

About