Microscopy Data Segmentation for Atomic-Scale Chemical Interface Positioning

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for measuring interfaces in materials observed by microscopy struggle to universally distinguish between chemical and structural interfaces and lack atomic spatial resolution, failing to accurately measure layer thicknesses, especially in ultrathin layers.

Innovation Solution

A method for processing multidimensional microscopy data involves normalization, segmentation, and thresholding to identify chemical interfaces, followed by rectification to achieve precise interface positioning and thickness measurement, utilizing machine learning and elastic deformation to handle hyperspaces.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If existing methods for measuring interfaces in materials are used, then general interface detection is possible, but measurement precision deteriorates due to inability to universally distinguish between chemical and structural interfaces and lack of atomic spatial resolution

Engineering Contradiction:
Improveinterface positioning precisionVSAvoidability to distinguish chemical and structural interfaces
Core Design Contradiction:
Measurement precisionVSAdaptability or versatility

Solution Approach 1:

The patent segments the interface detection process into distinct stages: (1) normalization of microscopy data, (2) segmentation into homogeneous zones, (3) calculation of standard deviation to identify interfaces, and (4) rectification for precise positioning. This segmentation allows the method to specifically target chemical interfaces through standardized processing steps while maintaining versatility across different material types and interface configurations.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent employs parameter changes by transforming the microscopy data through normalization (adjusting contrast to predetermined minimum and maximum values) and by calculating standard deviation as a new parameter to identify interfaces. These parameter transformations enable the method to achieve atomic spatial resolution and universally distinguish between chemical and structural interfaces by converting raw imaging data into standardized metrics suitable for precise measurement.

Inventive Principle:
Principle #35Parameter changes

2Manufacturing precision

If existing interface measurement methods are applied, then general material analysis is possible, but manufacturing precision deteriorates due to inaccurate layer thickness measurement especially in ultrathin layers

Engineering Contradiction:
Improvelayer thickness measurement accuracyVSAvoidultrathin layer measurement capability
Core Design Contradiction:
Manufacturing precisionVSMeasurement precision

Solution Approach 1:

The patent applies preliminary action by performing normalization of the microscopy data before interface detection. The normalization step (adjusting contrast to predetermined minimum and maximum values) prepares the data in advance to enhance the visibility and measurability of ultrathin layers, enabling accurate thickness measurement of layers as thin as 1 atomic monolayer even in corner regions.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent replaces traditional mechanical measurement approaches with computational methods. Instead of relying on direct physical measurement techniques that struggle with ultrathin layers, the method uses image processing algorithms (normalization, segmentation, standard deviation calculation, and rectification) to precisely determine layer thicknesses at the atomic scale, achieving manufacturing precision for ultrathin layers that was previously unattainable.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

3Measurement precision

If high-resolution microscopy imaging is used, then atomic spatial resolution is achieved, but device complexity increases due to the need for sophisticated data processing algorithms

Engineering Contradiction:
Improveatomic spatial resolutionVSAvoiddata processing algorithm complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the complex data processing into distinct, manageable modules: normalization module (contrast adjustment), segmentation module (identification of homogeneous zones), standard deviation calculation module (interface detection), and rectification module (precise positioning). This segmentation reduces the perceived complexity by breaking down the sophisticated algorithm into standardized processing steps that can be systematically implemented.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent simplifies the data processing complexity by transforming raw microscopy data into standardized parameters through normalization (converting to predetermined minimum and maximum contrast values) and by using standard statistical measures (standard deviation) for interface detection. These parameter transformations reduce the complexity of handling high-resolution data while maintaining atomic spatial resolution capability.

Inventive Principle:
Principle #35Parameter changes

4Productivity

If conventional interface detection methods are used, then general material characterization is possible, but productivity deteriorates due to inability to efficiently measure ultrathin layers and corners

Engineering Contradiction:
Improvemeasurement efficiencyVSAvoidultrathin layer and corner measurement accuracy
Core Design Contradiction:
ProductivityVSManufacturing precision

Solution Approach 1:

The patent ensures continuity of useful action by applying the same normalization and processing steps across the entire microscopy data field, including corner regions and ultrathin layers. The standardized algorithm processes all areas uniformly, maintaining measurement accuracy throughout the entire sample without requiring separate handling of different regions, thereby improving productivity while preserving precision for difficult-to-measure areas.

Inventive Principle:
Principle #20Continuity of useful action

Data Source

PatentEP4687100A1Method and device for processing multidimensional microscopy data for interface positioning between chemically homogeneous regions of a material sample
Publication Date: 2026.02.04 COMMISSARIAT A LENERGIE ATOMIQUE ET AUX ENERGIES ALTERNATIVES
  • EP4687100A1 patent drawingFigure 1
  • EP4687100A1 patent drawingFigure 2
  • EP4687100A1 patent drawingFigure 3

AI summary

This process includes acquiring at least one microscopy image of said sample forming an input data block, and the steps of: -normalizing (42) the input data block to obtain a normalized data block, including adjusting the contrast of the input data block; -segmenting (43) the normalized data block to determine a first data block representative of a first homogeneous zone and a second data block representative of a second homogeneous zone, such that the point-by-point sum of the values ​​of the first data block and the second data block is equal to the same value; -calculating (50) a standard deviation data block between said first and second data blocks, then thresholding (52) the standard deviation data block to obtain the chemical interface between the first homogeneous zone and the second homogeneous zone.