Large-scale time-lapse scanning electron microscopy image mosaic using a smooth stitching strategy.

Area of Science:

• Microscopy
• Image Processing
• Computational Science

Background:

• Microscopy often requires balancing field of view and resolution, necessitating the stitching of multiple high-resolution image tiles.
• Acquiring wide-view panoramic images from sequential scans is crucial for detailed analysis in various scientific applications.

Purpose of the Study:

• To develop an automatic image mosaic strategy for sequential 2D time-lapse scanning electron microscopy (SEM) images.
• To improve the accuracy and robustness of image stitching for large-area, high-resolution microscopy data.

Main Methods:

• An automatic strategy for computing pairwise translations between serial image tiles with indeterminate overlap.
• Feature point detection and matching limited by geographical coordinates to prevent mismatches.
• Incorporation of a smooth stitching field to manage nonlinear deformations and perspective transformations.

Main Results:

• The proposed method accurately computes translations and handles nonlinear deformations in image tiles.
• Experimental results show superior image stitching accuracy compared to existing mosaic algorithms.
• The strategy effectively avoids accidental feature point mismatches.

Conclusions:

• The developed automatic image mosaic strategy offers improved stitching accuracy for SEM images.
• This method has significant potential for high-resolution, large-area analysis using serial microscopy data.
• The approach addresses key challenges in panoramic image reconstruction from microscopy datasets.