Pyramid cluster detection and delineation by consensus

Pyramid algorithms for compact region detection and delineation can be used to detect compact dot clusters on a sparsely dotted background. When the delineation process is applied to a detected cluster, it yields 'ragged' results which are sensitive to the position of the cluster in the image. But if the process is applied to a set of shifted versions of the cluster, the results can be combined into an acceptable delineation. This illustrates a general 'consensus' approach to combining multiple techniques, or multiple versions of the same technique, to obtain more reliable results.