A novel workspace for image clustering

Abstract

A novel image clustering method based on the image histogram, which is processed by the empirical mode decomposition (EMD) is presented. An intermediate step derived from the EMD, which can decompose any nonlinear and non-stationary data into a number of intrinsic mode functions (IMFs) is exploited. The IMFs of the image histogram have interesting characteristics and provide a novel workspace that is utilized in order to automatically detect the different clusters into the image under examination. The proposed method was applied to several real and synthetic images and the obtained results show good image clustering robustness.