As color is a useful characteristic of our surrounding world, it gives clue for the recognition, indexing and retrieval of the images presenting the visual similarity. Thus, this paper focuses on the proper choice of the similarity measure applied to compare features evaluated in process the modeling of lossy coded color image information, based on the mixture approximation of chromaticity histogram. The analyzed similarity measure are those based on Kullback − Leibler Diverence, as Goldberger approximation and V ariational approximation. Signaturebased distance function as Hausdorff Distance, Perceptually Modified Hausdorff Distance and EarthMover's Distance were also investigated. Retrieval results were obtained for RGB, I1I2I3,Y UV, CIE XY Z, CIE L ∗ a ∗ b ∗, HSx, LSLM and TSL color spaces.
CITATION STYLE
Luszczkiewicz-Piatek, M. (2017). Image similarity in gaussian mixture model based image retrieval. In Advances in Intelligent Systems and Computing (Vol. 525, pp. 87–95). Springer Verlag. https://doi.org/10.1007/978-3-319-47274-4_10
Mendeley helps you to discover research relevant for your work.