Improved Image Segmentation Algorithm Using Graph-Edges

Abstract

In this paper an efficient algorithm for segment digital image has beendeveloped by measuring the evidence for a boundary between two regions in an imageusing (graph-edges). The regions in the image were sorted as components, where eachregion in an image represents a component in the graph. The region comparisonpredicate evaluates if there is evidence for a boundary between a pair of componentsby checking if the difference between the components, is large relative to the internaldifference within at least one of the components. A threshold function is used tocontrol the degree the difference between components must be larger than minimuminternal difference. An important characteristic of the method is its ability to preservedetail in important image regions while ignoring detail in unimportant regions. Theclassical methods depend just on external difference and ignore the internaldifference, when segment two neighboring regions.