A doubly degenerate diffusion model based on the gray level indicator for multiplicative noise removal

Area of Science:

• Image Processing
• Computational Imaging
• Applied Mathematics

Background:

• Multiplicative noise significantly degrades image quality, posing a challenge in image processing.
• Traditional variation methods have limitations in addressing multiplicative noise.
• Nonlinear diffusion models show promise for noise removal, particularly in additive noise scenarios.

Purpose of the Study:

• To develop a new framework for multiplicative noise removal based on nonlinear diffusion equation theories.
• To propose and analyze a doubly degenerate diffusion model for effective image denoising.
• To present an efficient numerical scheme for implementing the proposed denoising model.

Main Methods:

• A nonlinear diffusion filter framework integrating image gradient and gray-level information.
• Development of a doubly degenerate diffusion model tailored for multiplicative noise.
• Implementation using an efficient numerical scheme with stabilization by fast explicit diffusion.

Main Results:

• The proposed doubly degenerate diffusion model demonstrates effectiveness in multiplicative noise removal.
• The numerical scheme ensures efficient and stable implementation of the denoising model.
• Experimental results validate the superior performance of the proposed approach.

Conclusions:

• The nonlinear diffusion approach offers a powerful alternative to traditional methods for multiplicative noise removal.
• The doubly degenerate diffusion model provides a robust solution for enhancing image quality.
• The efficient numerical scheme facilitates practical application of the proposed denoising technique.