Real-Time External Control Combined with Image Post-Processing for Mitigating SEM Vibration Distortion.

Area of Science:

• Materials Science
• Microscopy Engineering
• Image Processing

Background:

• Scanning electron microscopes (SEMs) are vital for material characterization but are sensitive to vibrations.
• Environmental and internal vibrations degrade SEM image accuracy, and existing solutions have limitations.
• Passive isolation and image post-processing alone cannot fully correct vibration-induced distortions.

Purpose of the Study:

• To develop and validate a hybrid framework for mitigating vibration-induced distortion in SEM images.
• To combine real-time active hardware vibration suppression with advanced image post-processing techniques.
• To improve the accuracy and reliability of high-magnification SEM imaging.

Main Methods:

• A hybrid framework integrating real-time active hardware suppression and post-processing was developed.
• Fast Fourier Transform (FFT) was used to extract periodic vibration features and quantify scan line offset.
• Real-time inverse offset was applied during imaging, coupled with adaptive median filtering and Laplacian edge enhancement.

Main Results:

• The hybrid framework significantly reduced peak-to-peak vibration values by 39.4% and edge transition width by 91.7% at 100 kx magnification.
• No-reference image quality (NIQE) scores improved by 58.9%, indicating enhanced image quality.
• Periodic vibration distortion was effectively eliminated at both 50 kx and 100 kx magnifications, demonstrating consistent performance.

Conclusions:

• The proposed hybrid framework offers a robust solution for mitigating vibration impacts in SEM imaging.
• The strategy is universally applicable, easy to implement, and compatible with various vibration interferences without source localization.
• This work provides a valuable reference for anti-vibration optimization in SEMs and other microscopic techniques like TEM and AFM.