Achieve Computer Vision with 2D CA—An Introduction

Samuel Chen

Abstract

Humans can easily recognize and distinguish thousands of visual categories. The best computer algorithms achieve only a fraction of human performance in terms of both the number of classes recognized and the accuracy in distinguishing between those classes.

In this study, I have utilized two-dimensional cellular automata (2D CA) as image sensors. I have demonstrated that there exists a class of 2D CA that can filter horizontal, vertical, and 45-degree lines among other inputs.

I have shown in this study that instead of using traditional computer algorithms for tasks such as computer vision, one can use simple programs for such concrete applications.