4X4 Census Transform

TL;DR

This paper introduces a 4X4 Census Transform (4X4CT) that uses a larger, overlapping pixel kernel to improve image textural detail and contrast in computer vision applications.

Contribution

The paper presents a novel 4X4 Census Transform that extends the traditional 3X3 kernel to enhance image feature extraction.

Findings

Enhanced image textural crispness and contrast

Potential for more meaningful solutions in visual computing

Preliminary results show improved feature representation

Abstract

This paper proposes a 4X4 Census Transform (4X4CT) to encourage further research in computer vision and visual computing. Unlike the traditional 3X3 CT which uses a nine pixels kernel, the proposed 4X4CT uses a sixteen pixels kernel with four overlapped groups of 3X3 kernel size. In each overlapping group, a reference input pixel profits from its nearest eight pixels to produce an eight bits binary string convertible to a grayscale integer of the 4X4CT's output pixel. Preliminary experiments demonstrated more image textural crispness and contrast than the CT as well as alternativeness to enable meaningful solutions to be achieved.