Electron Microscope Threshold Calibration for Accurate Electron Counting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Transmission electron microscopes face challenges in accurately counting incident electrons due to noise interference and count loss when setting inappropriate thresholds during binarization processing, which affects the linearity and detective quantum efficiency of the images obtained.
Innovation Solution
An electron microscope system that includes an electronic optical system, a camera with an image sensor, and a computation unit that repeatedly sets tentative thresholds, acquires frame images under Poisson process conditions, binarizes and integrates them, and calculates normalized constants to determine an optimal threshold for binarization, ensuring accurate electron counting and improved linearity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a fixed threshold is used for binarization, then the processing is simple and fast, but noise is counted as incident electrons or incident electrons fail to be counted
Solution Approach 1:
The patent performs preliminary actions by acquiring multiple frame images under Poisson process conditions before determining the optimal threshold. The computation unit calculates normalized constants from these preliminary frames and uses them to establish an accurate threshold, ensuring that the subsequent binarization processing achieves high measurement precision without sacrificing processing efficiency.
Solution Approach 2:
The patent changes the threshold parameter dynamically based on calculated normalized constants rather than using a fixed threshold. By adjusting the threshold parameter according to the statistical properties of the acquired frames, the system achieves accurate electron counting while maintaining processing efficiency through automated parameter optimization.
2Productivity
If an inappropriate threshold is set during binarization, then processing is faster, but count loss occurs and linearity deteriorates
Solution Approach 1:
The patent implements feedback by calculating normalized constants from acquired frame images and using these calculations to determine the optimal threshold. This feedback mechanism ensures that the threshold setting is based on actual image data characteristics, maintaining both processing efficiency and image linearity through automated, data-driven threshold optimization.
Solution Approach 2:
The system performs preliminary analysis by acquiring multiple frames and calculating normalized constants before final threshold determination. This preliminary action ensures that the threshold is optimized for the specific imaging conditions, preserving linearity while maintaining processing efficiency through pre-computation of threshold parameters.
3Measurement precision
If multiple frames are acquired and processed to determine optimal threshold, then measurement precision is improved, but processing time increases
Solution Approach 1:
The patent performs threshold optimization as a preliminary calibration action using a set of reference frames. Although this requires additional processing time during calibration, it establishes an accurate threshold that can be reused for subsequent imaging, reducing the time loss to a one-time calibration cost rather than continuous processing overhead.
Solution Approach 2:
The system changes the processing approach by calculating normalized constants from multiple frames to determine an optimal threshold parameter. While this requires additional computational steps during calibration, it achieves superior measurement precision by base the threshold on statistical analysis of multiple frames rather than arbitrary fixed values.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The system effectively sets appropriate thresholds, reducing noise interference and count loss, thereby enhancing the linearity between pixel values and the number of incident electrons, and maintaining high detective quantum efficiency even when the image sensor is degraded.
Implementation Method 1
a camera (40) that includes an image sensor (42) and outputs a frame image which is based on a signal obtained by electrons entering each cell of the image sensor
Data Source
Figure 1~2
Figure 3~4
Figure 5
AI summary
An electron microscope (100) includes: an electronic optical system (20) that irradiates a specimen (S) with an electron beam and forms an image using electrons transmitted through the specimen (S); a camera (40) that includes an image sensor (42) and outputs a frame image which is based on a signal obtained by electrons entering each cell of the image sensor (42); and a computation unit (60) that generates an image based on the frame image, wherein the computation unit (60) performs processing for: setting a threshold; and binarizing the frame image using the threshold, and generating the image based on the binarized frame image, and in the processing for setting the threshold, the computation unit (60) repeatedly performs processing for (i) setting a tentative threshold, (ii) acquiring a plurality of the frame images obtained on a condition that electrons entering the image sensor follow Poisson process, (iii) binarizing each of the plurality of acquired frame images using the tentative threshold, (iv) generating an integrated image by integrating the plurality of binarized frame images, and (v) obtaining a normalized constant based on a mean and variance of pixel values of pixels of the integrated image, with the tentative threshold being differed each time of the processing, and the computation unit (60) obtains an optimal threshold based on the plurality of obtained normalized constants and sets the optimal threshold as the threshold.