The contrast-to-noise ratio for image quality evaluation in scanning electron microscopy.

Area of Science:

• Materials Science
• Microscopy
• Image Analysis

Background:

• Quantitative noise measurement is crucial for reliable scanning electron microscope (SEM) image analysis.
• The signal-to-noise ratio (SNR) is a common metric, but its applicability in certain SEM contexts is limited.
• Characterizing alternative metrics like the contrast-to-noise ratio (CNR) is essential for advancing SEM image quality assessment.

Purpose of the Study:

• To present and characterize the contrast-to-noise ratio (CNR) as a quantitative tool for noise measurement in SEM images.
• To analytically and experimentally investigate the analogies and differences between CNR and the widely used signal-to-noise ratio (SNR).
• To provide practical resources for implementing CNR in SEM image evaluation.

Main Methods:

• Analytical investigation comparing CNR and SNR.
• Experimental validation of CNR and SNR using SEM imaging.
• Development of a standard specimen and evaluation program for CNR application.

Main Results:

• CNR is characterized as an effective tool for quantitative noise measurement in SEM images.
• Key analogies and differences between CNR and SNR were identified through analysis and experimentation.
• A standard specimen and evaluation program facilitate practical CNR implementation.

Conclusions:

• CNR offers a valuable alternative or complement to SNR for noise assessment in SEM.
• The presented tools enable more robust and quantitative evaluation of SEM image quality.
• This work contributes to improved accuracy and reliability in SEM-based research and applications.