Automatic 1D Histogram Segmentation and Application to the Computation of Color Palettes

This article presents an implementation of the FTC (Fine-to-Coarse) algorithm for histogram segmentation, presented by Delon et al. in 2007. uses a non-parametric contrario approach to segment 1D into its meaningful modes. We describe also how method may be applied hue, saturation and intensity histograms color images order automatically extract their more representative colors, so-called palette. The palette extraction described this paper is based on one first published 2007 al., with improvement that affects low-saturated colors. Several results illustrate effectiveness algorithm.