Graph Based Semi and Unsupervised Classification and Segmentation of Microscopic Images

In this paper, we propose a general formulation of discrete functional regularization on weighted graphs. This framework can be used on any multi-dimensional data living on graphs of the arbitrary topologies. In this work, we focus on microscopic image segmentation and classification within semi and unsupervised schemes. Moreover, to provide a fast image segmentation we propose a graph based image simplification as a pre-processing step. Biological elements contained in images such as cells, cytoplasm and nuclei are segmented and classified with this image simplification and label diffusion processes on weighted graphs.