Dynamical Threshold-Based Fractional Anisotropic Diffusion for Speckle Noise Removal

Jiali Wei , Xiaofeng Liao

IEEE Transactions on Image Processing : a Publication of the IEEE Signal Processing Society +

|

April 22, 2025

View abstract on PubMed

Summary

This study introduces a novel dynamical threshold-based fractional anisotropic diffusion (DTFAD) method for effective image speckle noise removal. DTFAD outperforms traditional methods by preserving image features and textures.

Area Of Science

Image Processing
Computational Imaging
Signal Processing

Background

Speckle noise is a challenging multiplicative noise in image processing.
Its intensity dependence on the signal makes removal difficult.
Traditional methods struggle with preserving image details.

Purpose Of The Study

To develop a novel approach for effective speckle noise removal.
To enhance image denoising while preserving fundamental features and edges.
To introduce the dynamical threshold-based fractional anisotropic diffusion (DTFAD) model.

Main Methods

Utilizing fractional derivative integrated with anisotropic diffusion.
Incorporating gradient and gray scale image information.
Implementing a dynamic threshold function for adaptive diffusion.
Employing an explicit finite difference scheme for model implementation.

Main Results

The DTFAD model demonstrates superior speckle noise removal compared to traditional anisotropic diffusion.
Achieves a better balance between denoising performance and texture preservation.
Well-posedness of the DTFAD model is established.

Conclusions

The DTFAD model offers an effective solution for speckle noise reduction.
It shows significant potential for practical engineering applications.
Preserves essential image features and textures during denoising.