HIERARCHICAL IMAGE-ANALYSIS USING IRREGULAR TESSELLATIONS

We present a novel multiresolution image analysis technique based on hierarchies of irregular tessellations generated in parallel by independent stochastic processes. Like the "traditional" image pyramids these hierarchies are constructed in on the order of log(image-size) steps. However, the structure of a hierarchy is adapted to the image content and artifacts of rigid resolution reduction are avoided. We give two applications of our technique: connected component analysis of labeled images, and segmentation of gray level images. In labeled images, every connected component is reduced to a separate root, with the adjacency relations among the components also extracted. In gray level images the output is a segmentation of the image into a small number of classes as well as the adjaceny graph of the classes.