Monogenic Signal Associated with Linear Canonical Transform and Application to Edge Detection Problems

TL;DR

This paper extends the monogenic signal concept to high-dimensional spaces using the Riesz transform, analyzing local phase and attenuation for improved edge detection.

Contribution

It introduces a high-dimensional monogenic signal framework and applies differential phase congruency for edge detection tasks.

Findings

Effective high-dimensional feature representation

Enhanced edge detection performance

Analysis of local phase and attenuation

Abstract

Monogenic signal is regarded as a generalization of analytic signal from the one dimensional space to the high dimensional space. It is defined by an original signal with the combination of Riesz transform. Then it provides the signal features representation, such as the local attenuation and the local phase vector. The main objective of this study is to analyze the local phase vector and the local attenuation in the high dimensional spaces. The differential phase congruency is applied for the edge detection problems.