Component-Trees and Multi-value Images: A Comparative Study

Abstract

In this article, we discuss the way to derive connected operators based on the component-tree concept and devoted to multi-value images. In order to do so, we first extend the grey-level definition of the component-tree to the multivalue case. Then, we compare some possible strategies for colour image processing based on component-trees in two application fields: colour image filtering and colour document binarisation.

Component-Trees and Multivalued Images: Structural Properties

“…After a preliminary study of the relations between component-trees and multivalued images from a methodological point of view [1], a generalisation of component-trees to such images has been initiated in [2]. The present work develops this framework.…”

Color processing using max-trees: A comparison on image compression

“…They allow users to chose properties of sections of the image that can be ignored, over-processed or filtered out. Connected filters have been used in various application, including image filtering and noise reduction [6], [7], [8], image simplification for compression [9], [10], video processing [10], [2], vessel enhancement filtering [11], [12], [13], and image analyzed microscopy [14]. Recent reviews can be found in [15], [16].…”

“…Both are also referred to as component trees [17], [18], [8]. The aim of these trees is to encode a hierarchy of connected components at different levels to allow fast filtering or for analysis as a multi-scale representation of the image or volume under study.…”

“…This poses problems in component-tree-based processing, because the hierarchy in the tree is driven by the total order of the pixel values. To build such trees in color morphology, the ordering has to be decided upon [21], [6], [8].…”

“…A marginal ordering deals with each component independently and then later concatenates the scalars back together. This method has been shown to introduce new colors [8]. Reduced ordering obtains a scalar value from the vectorial components.…”