TEM Camera Outlier Removal via Sub-Frame Statistical Thresholding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current electron microscopy image processing techniques fail to effectively eliminate radiation-induced spots, particularly x-ray-generated outliers, which degrade image quality due to reliance on ergodicity assumptions and inadequate shielding, especially in high-contrast or rapidly varying specimens, and are exacerbated by increasing beam intensity.
Innovation Solution
The method involves fractionating the exposure into sub-exposures to allow pixel-wise evaluation of mean and standard deviation, reducing the dose for thresholding, and iteratively correcting outliers, which reduces artifacts and improves outlier removal by making statistical evaluations independent of specimen characteristics and radiation signal overlap with the indirect image signal.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If local image statistics are evaluated to establish a local expected range for outlier removal, then radiation-induced spots can be identified and removed, but high-contrast features in images (such as diffraction pattern spots) are falsely identified as radiation event outliers
Solution Approach 1:
The patent segments the image processing into multiple passes: first identifying potential outliers using statistical thresholds, then verifying them through additional criteria before removal. This multi-stage segmentation approach prevents false identification of high-contrast features as outliers while maintaining effective radiation spot removal
Solution Approach 2:
The patent implements feedback mechanisms where the outlier detection process uses information from previously processed pixels and adjusts thresholds dynamically. The system learns from the image characteristics and refines its outlier identification, reducing false positives while maintaining reliability in radiation spot detection
2Object-affected harmful factors
If the mass between scintillator and sensor is increased to reduce x-ray spots, then radiation shielding is improved, but the device complexity and space requirements increase
Solution Approach 1:
The patent replaces the mechanical approach of adding physical mass for radiation shielding with an electronic/image processing solution. By using software-based outlier detection and removal algorithms, the system eliminates radiation-induced spots without requiring additional physical shielding materials, thereby reducing device complexity and space requirements
Solution Approach 2:
The patent changes the approach from modifying physical parameters (adding mass) to modifying processing parameters (image analysis thresholds and algorithms). By adjusting detection sensitivity and using statistical methods, the system achieves radiation spot removal without changing the physical structure of the camera
3Illumination intensity
If the electron beam intensity is increased to improve image signal, then image brightness is improved, but radiation-induced spots increase in frequency
Solution Approach 1:
The patent converts the harmful effect of increased radiation spots (which naturally occurs with higher beam intensity) into a detectable pattern. By using statistical analysis to identify spots that deviate from expected distributions, the system can distinguish radiation-induced artifacts from genuine image features, allowing high beam intensity to be used while maintaining image quality through post-processing correction
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
This approach significantly reduces radiation-induced artifacts, improves image quality by accurately identifying and correcting outliers, and maintains effectiveness across varying specimen contrast and beam intensities, enhancing the precision of image processing in electron microscopy.
Implementation Method 1
employing either a scintillator to convert an electron image into a light image
Implementation Method 2
a silicon sensor to capture the light image
Implementation Method 3
x-rays generated through bremsstrahlung by the beam at the scintillator
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Methods are disclosed for removal of outlier pixels from a transmission electron microscopy camera image. One exemplary method includes establishing a desired exposure of n electrons per pixel; exposing the camera to a series of sub-frame exposures to produce a series of sub-frame images; calculating an average image signal of all sub-frame exposures in said series; establishing a threshold selected to achieve a desired number of false positives; evaluating each of said sub-frame exposures for pixels further away from said average than said threshold; and replacing pixels in each of said sub-frame images that exceed said threshold with said average to form corrected sub-frame images.