6533b855fe1ef96bd12b10a7

RESEARCH PRODUCT

Advances in the statistical methodology for the selection of image descriptors for visual pattern representation and classification

Pavel PudilJana NovovičováJosef KittlerFrancesc J. Ferri

subject

Computer sciencebusiness.industryVisual descriptorsVisual patternsRepresentation (systemics)A priori and a posterioriPattern recognitionArtificial intelligencebusinessMachine learningcomputer.software_genrecomputerSelection (genetic algorithm)

description

Recent advances in the statistical methodology for selecting optimal subsets of features (image descriptors) for visual pattern representation and classification are presented. The paper attempts to provide a guideline about which approach to choose with respect to the a priori knowledge of the problem. Two basic approaches are reviewed and the conditions under which they should be used are specified. References to more detailed material about each one of the methods are given and experimental results supporting the main conclusions are briefly outlined.

yearjournalcountryeditionlanguage
1995-01-01
https://doi.org/10.1007/3-540-60268-2_389