Image segmenting apparatus and method

Abstract

An image segmenting apparatus and method is provided. The image segmenting apparatus includes an initial image segmenting unit, a region structurizing unit and a redundant region combiner. The initial image segmenting unit converts color signals of an input image into a color space which is based on predetermined signals, and segments the input image into a plurality of regions according to positions of color pixels of the input image in the color space. The region structurizing unit classifies the plurality of regions into layers according to horizontal, adjacent relation and hierarchical, inclusive relation between the regions, and groups adjacent regions into region groups in each layer, so as to derive a hierarchical, inclusive relation between the region groups. The redundant region combiner determines the order in which adjacent regions are combined according to the horizontal, adjacent relation between regions and the hierarchical, inclusive relation between region groups. The redundant region combiner also determines whether to combine adjacent regions according to the determined combination order, and combines adjacent regions if the adjacent regions are determined to be substantially the same. Even if regions appears to be adjacent each other in a region adjacent graph (RAG), a structural inclusive relation between regions can be derived by excluding the combination of the regions or rearranging their combination order according to a hierarchical structure. Subsequently, the mutual relation between two regions can be inferred from the inclusive relation even if the color signals of the two regions, for example, a region in a highlighted area and a region in its surrounding area, are not similar to each other.