• Framework

  • Motivation

                    (a) Original image, (b) sequentially filtered image, (c) binary image, (d) distance image, and (e) topographic surface

                    In our work, we observe that the distance image of overlapped nuclei shows a Gaussian mixture distribution shown in (e).

  • Experimental results

      • Chanho Jung, Changick Kim, S. W. Chae, and S. Oh, "Unsupervised Segmentation of Overlapped Nuclei Using Bayesian Classification," IEEE Transactions on Biomedical Engineering, vol. 57, no. 12, pp. 2855-2832, Dec. 2010.