A pyramidal approach for the recognition of neurons using key features

摘要

Pyramid-like hierarchical structures have been shown to be suitable for many computer vision tasks. In this paper a pyramidal approach for recognition of neurons from an electron micrograph is described. Key features for object recognition are defined. Transforms for extracting features are performed hierarchically in a simulated pyramid, cascading local parallel operations. The algorithms described in this paper are expected to be fairly expandable to general purpose pyramidal vision systems.