Figure - PMC

Skip to main content

An official website of the United States government

Here's how you know

Here's how you know

Official websites use .gov
A .gov website belongs to an official government organization in the United States.

Secure .gov websites use HTTPS
A lock ( ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

View full-text article in PMC

. 2019 May 13;15(5):e1007012. doi: 10.1371/journal.pcbi.1007012

Search in PMC
Search in PubMed
View in NLM Catalog
Add to search

© 2019 Kulikov et al

This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

PMC Copyright notice

Fig 2 — (a) The architecture of shallow DoGNet. The input image channels (for example synapsin, vGlut, and PSD95) are each processed by five trainable DoG filters. The weighted sum (with trainable weights) combines the resulting 15 DoG layer output maps into a single map. The sigmoid function converts the latter map into a pixel probability map. (b,c,d) The variations of the Difference of Gaussians that we use in each DoG layer. (b) An isotropic Difference of Gaussians. (c) An anisotropic difference of Gaussians. Each Gaussian is described by a pair of variance values and a rotation angle. (d) A 3D Isotropic Difference of Gaussians. Surfaces show filter values along z slices.