The Main Cognitive Model of Visual Recognition: Contour Recognition

TL;DR

This paper develops a comprehensive cognitive model for visual recognition focusing on contour detection on 3D surface patterns, refining previous models to better reflect essential recognition characteristics.

Contribution

It introduces a complete and refined cognitive model for recognizing patterns with contours on 3D surfaces, improving upon earlier, coarser models.

Findings

Revealed inherent laws of pattern recognition involving contours.

Established a detailed cognitive model reflecting recognition characteristics.

Corrected and expanded previous models for better accuracy.

Abstract

In this paper, we will study the following pattern recognition problem: Every pattern is a 3-dimensional graph, its surface can be split up into some regions, every region is composed of the pixels with the approximately same colour value and the approximately same depth value that is distance to eyes, and there may also be some contours, e.g., literal contours, on a surface of every pattern. For this problem we reveal the inherent laws. Moreover, we establish a cognitive model to reflect the essential characteristics of the recognition of this type of patterns. In [1], a coarser model or a basicer one is described. In this paper, some important errors are revised, some key things are added, at last, a complete model is described.