Figure 2: Illustrative examples for voting-based methods.
From: Multi-Pass Adaptive Voting for Nuclei Detection in Histopathological Images
(a) A synthetic nucleus with a few inverse gradient directions. (b) Illustration of region with a large number of votes. (c) An example of poorly stained tissue image with its edge map and edge fragments for one of the nuclei with the inverse gradients is shown. (d) An illustrative example of a nucleus with a hollow interior. (e) Examples of clustered nuclei. The thick blue arrows reflect the gradient information on the inner edge of the nuclei. Note that the outwards gradients are easily lead to potential false positive detections with voting based approaches.
