On Multichannel Neurons, with An Application to Template Search

Authors

  • Helio Perroni Filho Universidade Federal do Esp´ırito Santo, Laboratorio de Computac¸ ´ ao de Alto Desempenho
  • Alberto Ferreira De Souza Universidade Federal do Esp´ırito Santo, Laboratorio de Computac¸ ´ ao de Alto Desempenho

Keywords:

cognitive models, computer vision, fourier transform, GUI automation, machine learning, template matching

Abstract

The multichannel model is a complete reassessment of how neurons work at the biochemical level. Its results can be extended into an overarching theory of how vision, memory and cognition come to be in the living brain. This article documents a first attempt at testing the model’s validity, by applying its principles to the construction of an image templatematching framework, which is then used to solve a Graphical User Interface (GUI) automation problem. It was found that the template-matching function thus implemented can consistently locate required visual controls, even when template and match differ by color palette or (to a slighter degree) feature
proportion. The article concludes by discussing the significance of these results and directions for further research.

Downloads

Download data is not yet available.

Downloads

Published

2014-01-01

How to Cite

Helio Perroni Filho, & Alberto Ferreira De Souza. (2014). On Multichannel Neurons, with An Application to Template Search. Journal of Network and Innovative Computing, 2, 12. Retrieved from https://cspub-jnic.org/index.php/jnic/article/view/46

Issue

Section

Original Article