An expanded approach to noise reduction from high-resolution STEM images based on the maximum entropy method

Area of Science:

• Materials Science
• Microscopy
• Image Processing

Background:

• High-angle annular dark-field scanning transmission electron microscopy (HAADF STEM) is crucial for high-resolution imaging.
• Noise in experimental HAADF STEM images can obscure fine details and hinder analysis.
• Existing noise reduction methods may be complex or less effective for high-resolution data.

Purpose of the Study:

• To propose an expanded application of the maximum entropy method (MEM) for noise reduction in HAADF STEM images.
• To develop a streamlined noise reduction technique with minimal user-defined parameters.
• To demonstrate the efficacy of the proposed method in denoising high-resolution HAADF STEM images.

Main Methods:

• The maximum entropy method (MEM) is adapted and expanded for noise reduction.
• The MEM is integrated with noise standard deviation estimation from experimental HAADF STEM images.
• Low-pass filtering is applied using the information limit for incoherent STEM imaging.

Main Results:

• The developed method effectively reduces noise in high-resolution HAADF STEM images.
• The technique requires only one parameter, a Lagrange multiplier, simplifying its application.
• The noise reduction is achieved without significant loss of image resolution.

Conclusions:

• The expanded MEM offers an efficient and user-friendly approach for denoising HAADF STEM images.
• This method enhances the quality of high-resolution STEM imaging, facilitating more accurate analysis.
• The simplified parameterization makes the technique broadly applicable in electron microscopy.