A set of neural lattices that use the central limit for Fourier and Gabor transforms, multiple-scale Gaussian smoothing, and edge detection

摘要

A set of neural lattices based on the central limit theorem is described. Each of the described lattices generates in parallel a set of multiscale Gaussian smoothings of their input arrays. The recursive smoothing principle of the lattices can be extended to any dimension. In addition, the lattices can generate a variety of multiple-scale operators such as the edge detectors of J. Canny (1986), Laplacians of Gaussians, and multidimensional Fourier and Gabor transforms.