APPARATUS FOR DETECTING COLOR REGION USING GAUSSIAN MIXTURE MODEL AND ITS METHOD

摘要

The present invention relates to an apparatus and method for detecting a color region using a Gaussian mixture model and includes a first feature vector extraction unit for extracting a first color feature vector using a color space transformation from a reference image, A color distribution modeling unit for modeling a color distribution of a reference image by calculating a Gaussian mixture model using an EM algorithm using a color feature vector; and a second feature extracting unit for extracting a second color feature vector using an input color space transformation And a color area detecting unit for detecting a color area of the input image corresponding to the modeled color distribution using the extracted second color feature vector in units of pixels through a Gaussian mixture model, It is possible to effectively detect multiple color areas as well as one color area.