Automatic Astigmatism Correction in Electron Microscopes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing electron microscope image acquisition is hindered by astigmatism, leading to image artifacts and loss of fine details due to misaligned electron beams, which is difficult to correct using manual and visual methods.
Innovation Solution
An automatic image-based method for correcting astigmatism in electron microscopes by analyzing Fourier spectrum images to determine the most round distribution of pixels, adjusting stigmator settings to minimize elongation, and iteratively optimizing x- and y-stigmator settings using gray-weighted moments and vector ratios.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If manual and visual methods are used to correct astigmatism, then the correction process can be performed, but the process is complicated and cumbersome, reducing efficiency
Solution Approach 1:
The patent replaces manual visual inspection and mechanical adjustment methods with an automated image processing system that uses Fourier transform algorithms to detect and correct astigmatism. The calculating device automatically analyzes image data and determines optimal stigmator settings, eliminating the need for operators to manually assess images and adjust controls based on visual judgment.
Solution Approach 2:
The system performs self-correction by automatically analyzing the captured image, calculating the Fourier spectrum, identifying astigmatism characteristics through vector analysis, and determining the optimal stigmator settings without requiring external manual intervention. The calculating device serves itself by using the image data to guide the correction process.
2Measurement precision
If physical features of the microscope are used for alignment, then alignment can be performed, but the method relies on complicated and cumbersome features of the microscope itself
Solution Approach 1:
The patent extracts the essential alignment information from the image data itself rather than relying on physical alignment features of the microscope. By using Fourier transform analysis on the captured image, the system extracts astigmatism characteristics directly from the image pattern, separating the measurement function from the physical microscope components.
Solution Approach 2:
The system creates a mathematical representation (Fourier spectrum) of the image data to analyze alignment characteristics. Instead of directly manipulating physical microscope features, the calculating device works with a copied and transformed version of the image data, making the alignment process independent of complex physical features.
3Reliability
If astigmatism is not corrected, then the alignment process is faster, but image quality deteriorates with artifacts and loss of fine details
Solution Approach 1:
The patent implements continuous automated correction by integrating the Fourier transform analysis and stigmator optimization into the image acquisition process. The system continuously monitors image quality through Fourier analysis and automatically adjusts stigmator settings to maintain optimal image quality without interrupting the workflow for manual corrections.
Solution Approach 2:
The system uses feedback from the Fourier spectrum analysis to automatically determine optimal stigmator settings. The calculating device analyzes the Fourier spectrum to identify astigmatism characteristics, then uses this feedback information to adjust the stigmator settings, creating a closed-loop control system that maintains image quality.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The method is for automatic astigmatism correction of a lens system. A first image (96) is provided that is not in focus at a first stigmator setting of a set of lenses. A calculating device calculates a corresponding first Fourier spectrum image (312). A distribution and direction of pixels of the Fourier spectrum image (128, 130, 312) are determined by calculating a first vector (132) and a second vector (134). The first vector (132) is compared with the second vector (134). The lens system is changed from a first stigmator setting to a second stigmator setting to provide a second image (98). A corresponding Fourier spectrum image (314) is calculated. The distribution and direction of pixels of the second Fourier spectrum image (314) is determined by calculating a third vector and a fourth vector. The third vector is compared to the fourth vector. The image that has the lowest vector ratio is selected.