Using the NADEL to solve a problem in biomedical imaging
Jian Wu, Michael A. Penna · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · 1992
The correspondence problem is the problem of establishing a one-to-one correspondence between the same physical features in multiple images of a scene. Solving this problem is an important task that arises in the computer automation of many visual processes. In this paper, we discuss how the neural analog diffusion/enhancement layer (or NADEL) can be used to address the correspondence problem. Implementation results are presented in the context of a real problem in biomedical image processing: construction of 3-dimensional models of cytoskeleta organelles in a nerve fiber axon from 2-dimensional cross-sectional images of the axon.