Edge Detection Methods Based on Differential Phase Congruency of Monogenic Image

TL;DR

This paper introduces novel edge detection algorithms based on differential phase congruency in monogenic images, utilizing monogenic scale-space and Clifford analysis to improve image analysis in higher dimensions.

Contribution

It proposes new edge detection methods leveraging monogenic scale-space and differential phase congruency, enhancing understanding of higher-dimensional image analysis.

Findings

Effective edge detection demonstrated on medical images

Improved understanding of local phase and amplitude relations

Algorithms based on monogenic analysis outperform traditional methods

Abstract

Edge Detection Methods Based on Differential Phase Congruency of Monogenic Image Abstract: Edge detection has been widely used in medical image processing and automatic diagnosis. Some novel edge detection algorithms,based on the monogenic scale-space, are proposed by detecting points of local extrema in local amplitude, the local attenuation and modified differential phase congruency methods. The monogenic scale-space is obtained from a known image by Poisson and conjugate Poisson filtering. In mathematics, it is the Hardy space in the upper half-space. The boundary value of the monogenic scale-space representation is a monogenic image. In the monogenic scale-space, the definitions involving scale, such as local amplitude,local attenuation, local phase angle, local phase vector and local frequency (phase derivatives) are proposed. Using Clifford analysis, the relations between the local attenuation and the local phase vector are obtained. These study will be improved the understanding of image analysis in higher dimensional spaces. Experimental results are shown by using some typical images.