Gray-scale and colour image segmentation via region growing and region merging

This paper introduces image segmentation as an important first task of any image-analysis process. A seeded region-growing and -merging algorithm is created to segment gray-scale and videophone-type colour images. The approach starts with a set of seed pixels, and from these grows regions by appending to each seed pixel those neighbouring pixels that satisfy a certain predicate. Small regions of far-away values are merged into neighbouring regions. Regions of similar value are also merged. Homogeneity functions are introduced for both gray-scale and colour images. The colour-image segmentation approach utilizes the HSI (hue, saturation, intensity) colour space because of its close relation to the way in which people describe the perception of colour.