"Doktorandské dny 2007" - Search Results - Digital Repository

guest :: login Digital Repository
		Search		Submit		Help		About

Home > Search Results: "Doktorandské dny 2007"

Search:

Search Tips :: Advanced Search

Search collections:

Sort by:	Display results:	Output format:

	Informační shlukování kategoriálních dat Hora, Jan The EM algorithm has been used repeatedly to identify latent classes in categorical data by estimating finite distribution mixtures of product components. Unfortunately, the underlying mixtures are not uniquely identifiable and, moreover, the estimated mixture parameters are starting-point dependent. For this reason we use the latent class model only to define a set of ``elementary'' classes by estimating a mixture of a large number components. As such a mixture we use also an optimally smoothed kernel estimate. We propose a hierarchical ``bottom up'' cluster analysis based on unifying the elementary latent classes sequentially. The clustering procedure is controlled by minimum information loss criterion. Detailed record
	Near-periodic texture synthesis Hatka, Martin This paper describes two methods for seamless enlargement of difficult colour textures containing both regular periodic and stochastic components. Such textures cannot be modelled using neither simple tiling nor using purely stochastic models. The first novel method automatically recognizes and separates periodic and random texture components. Each of this components is subsequently modelled using the coresponding optimal method. Both independently enlarged texture components are combined in the resulting synthetic near regular texture. The second method detects two main direction of periodicity in regular periodic component and generates several double-toroidal tiles of the same general shape, which can seamlesly enlarge given near-periodic texture without visible regularity. Detailed record
	Obrazová databáze navržená pro rychlé a robustní vyhledávání snímků Horáček, Ondřej ; Flusser, Jan The task of the paper is to find originals of modified images, tipicallz stolen and republished on the web. Our problem is specific in terms of the database size millions of photos), demanded speed of the search (seconds), and unknown image modifications (loss of quality, radiometric degradation, etc.). The proposed method is based on image preprocess for normalization, retrieval of candidates from image index, and the candidate verification. Based on wide experiments, it was shown that the success rate of the search depends on the level of image modification. Detailed record

Interested in being notified about new results for this query?
Subscribe to the RSS feed.

Digital Repository :: :: :: ::
Powered by v1.1.2
Maintained by

This site is also available in the following languages:
Česky English