A general image misalignment correction method for tomography experiments

Area of Science:

• Imaging Science
• Materials Science
• Physics

Background:

• Tomography reconstructs 3D slices from 2D projections.
• Mechanical system offsets cause projection misalignment, reducing spatial resolution.
• Existing methods lack efficiency and accuracy in correcting these misalignments.

Purpose of the Study:

• To present a novel and effective method for correcting image misalignments in tomography.
• To improve the spatial resolution and accuracy of 3D reconstructed images.
• To develop a generic method applicable to various tomography techniques.

Main Methods:

• Introduced outer contour-based misalignment correction (OCMC).
• Utilized the sample's outer contour as auxiliary information for misalignment estimation.
• Developed an end-to-end correction method for seamless integration.

Main Results:

• OCMC demonstrated significant improvements in alignment accuracy.
• The method showed high time efficiency compared to existing techniques.
• Validated across five diverse datasets and multiple tomography techniques.

Conclusions:

• OCMC effectively corrects image misalignments in tomography.
• The method enhances 3D reconstruction quality and spatial resolution.
• OCMC is suitable for online processing and future feedback control in experiments.