Auto Sweep ROI Selection for Focus and Astigmatism Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional charged-particle beam (CPB) imaging systems face inefficiencies in automatic focus and astigmatism correction due to sweeping the entire image field of view, which is slow and requires manual operator intervention for ROI selection, leading to inconsistency and additional adjustments when samples move or rotate.
Innovation Solution
The method involves producing gradient-based images to identify candidate ROIs, evaluating them using eigenvalues and histograms of gradients, and ranking these regions based on image pyramids to automate the selection process, ensuring suitable regions are identified for focus and astigmatism correction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If the entire image field of view is used for automatic focus and astigmatism correction, then the correction can be performed over a complete area, but the execution time becomes slow
Solution Approach 1:
The patent divides the entire image field of view into multiple candidate regions of interest (ROIs) and evaluates them individually using gradient magnitude and eigenvalue criteria. This segmentation allows the system to identify and use only the most suitable regions for correction, rather than processing the entire FOV, thereby reducing execution time while maintaining correction quality.
Solution Approach 2:
The patent extracts specific candidate ROIs from the full image that possess optimal characteristics for focus and astigmatism correction. By using gradient-based identification and eigenvalue evaluation, the system extracts only the most informative regions, eliminating the need to process irrelevant areas and thus reducing processing time.
2Manufacturing precision
If manual ROI selection is performed by operators, then suitable regions can be identified, but consistency among users is difficult to achieve and additional manual adjustments are needed when samples move or rotate
Solution Approach 1:
The patent implements an automated ROI selection system that performs gradient magnitude calculation, candidate identification, and eigenvalue-based evaluation without human intervention. The algorithm independently identifies suitable ROIs and ranks them, eliminating user variability and providing consistent results across different operators and sample positions.
Solution Approach 2:
The patent uses objective mathematical parameters (gradient magnitude thresholds and eigenvalue criteria) to evaluate and select ROIs, replacing subjective manual judgment. These quantitative parameters provide consistent, repeatable selection criteria that automatically adapt to different samples and conditions, ensuring operational consistency.
3Manufacturing precision
If manual ROI selection is performed, then regions suitable for correction can be identified, but the process requires users with training and experience in imaging physics
Solution Approach 1:
The patent replaces the manual expert-based selection process with an automated computational algorithm. The system uses gradient-based image processing and eigenvalue analysis to objectively evaluate regions, substituting human expertise with mathematical computations that are easier to implement and maintain while achieving comparable or superior selection quality.
4Area of stationary object
If the entire image field of view is swept, then complete coverage is achieved, but the automatic focus and astigmatism correction functions execute slowly
Solution Approach 1:
The patent segments the full image into multiple candidate ROIs and evaluates them using gradient magnitude and eigenvalue criteria. By identifying and using only the most suitable candidate regions rather than sweeping the entire FOV, the system maintains adequate coverage for correction while significantly reducing the processing area and execution time.
Solution Approach 2:
The patent applies partial action by selecting and processing only a subset of the most suitable ROIs from the full image, rather than processing the entire FOV. This approach provides sufficient coverage for accurate focus and astigmatism correction while minimizing processing time and computational resources.
Data Source
AI summary
Sample regions of interest (ROIs) for use in autofocus procedures are identified based on a gradient image of the sample. ROIs with gradient values greater that a threshold are selected, and eigenvalues of the associated image matrices are determined. ROIs with suitable variation in eigenvalues such as at least two relatively large eigenvalues are associated with high contrast features orientated in multiple directions so that such ROIs are suitable for automatic focus and astigmatism correction. Suitable ROIs can also be identified based on a histogram of gradient orientations.