Grain-based minerology segmentation system and method

Patent metadata

Field	Value
Publication number	US-12469104-B2
Application number	US-202017755162-A
Country	US
Kind code	B2
Filing date	Oct 23, 2020
Priority date	Oct 24, 2019
Publication date	Nov 11, 2025
Grant date	Nov 11, 2025

Abstract

A method of enhancing a resolution of an EDS image of a sample includes generating an EDS image of the sample, generating a non-EDS image of the sample generating, using a machine learning algorithm, an enhanced resolution EDS image of the sample based on the generated feature map and based on the first EDS image, where a resolution of the enhanced resolution EDS image is higher than a resolution of the first EDS image.

First claim

The invention claimed is:

1 . A method of automatically identifying mineral content of a sample, the method comprising:
generating a light microscopy image of the sample; identifying individual grains in the light microscopy image using a first clustering algorithm to identify grain boundaries on a per-pixel basis; identifying pore spaces in the sample based on output of a second clustering algorithm that analyzes the light microscopy image on a per-pixel basis, wherein the pore spaces are filled with a polymer material that includes epoxy material that has a uniform optical signal in the light microscopy image in comparison to optical signals from the individual grains identified in the light microscopy image, wherein identifying the individual grains in the light microscopy image using the first clustering algorithm is based, at least in part, on the identified pore spaces; classifying the identified individual grains in the light microscopy image using a third clustering algorithm; and mapping the classified individual grains to known mineral content.

2 . The method of claim 1 , wherein the light microscopy image includes a plurality of registered light microscopy images.

3 . The method of claim 1 , wherein classification of the identified individual grains by the third clustering algorithm includes classifying each grain based on aggregated statistics of a plurality of pixels corresponding to the grain.

4 . The method of claim 3 , wherein classifying each grain based on aggregated statistics of a plurality of pixels corresponding to the grain includes classifying each grain based on a distribution of pixel values corresponding to the grain.

5 . The method of claim 3 , wherein classifying each grain based on aggregated statistics of a plurality of pixels corresponding to the grain includes classifying each grain based on a distribution of filtered pixel values corresponding to the grain.

6 . A system for automatically identifying mineral content of a sample, the system comprising:
a processor; and a memory including executable instructions that when executed by the processor cause the processor to: receive a light microscopy image of the sample; identify individual grains in the light microscopy image using a first clustering algorithm to identify grain boundaries on a per-pixel basis; identify pore spaces in the sample based on output of a second clustering algorithm that analyzes the light microscopy image on a per-pixel basis, wherein the pore spaces are filled with a polymer material that includes epoxy material that has a uniform optical signal in the light microscopy image in comparison to optical signals from the individual grains identified in the light microscopy image, wherein identifying the individual grains in the light microscopy image using the first clustering algorithm is based, at least in part, on the identified pore spaces; classify the identified individual grains in the light microscopy image using a third clustering algorithm; and map the classified individual grains to known mineral content.

7 . The system of claim 6 , wherein the executable instructions that when executed by the processor further cause the processor to:
identify pore spaces in the sample based on output of a third clustering algorithm that analyzes the light microscopy image on a per-pixel basis, and wherein identifying the individual grains in the light microscopy image using the first clustering algorithm is based, at least in part, on the identified pore spaces.

8 . The system of claim 7 , wherein the pore spaces are filled with a polymer material.

9 . The system of claim 8 , wherein the polymer material includes epoxy material that has a uniform optical signal in the light microscopy image in comparison to optical signals from the individual grains identified in the light microscopy image.

10 . The system of claim 6 , wherein the light microscopy image includes a plurality of registered light microscopy images.

11 . The system of claim 6 , wherein classification of the identified individual grains by the third clustering algorithm includes classifying each grain based on aggregated statistics of a plurality of pixels corresponding to the grain.

12 . The system of claim 11 , wherein classifying each grain based on aggregated statistics of a plurality of pixels corresponding to the grain includes classifying each grain based on a distribution of pixel values corresponding to the grain.

13 . The system of claim 11 , wherein classifying each grain based on aggregated statistics of a plurality of pixels corresponding to the grain includes classifying each grain based on a distribution of pixel values corresponding to the grain.

14 . The system of claim 11 , wherein classifying each grain based on aggregated statistics of a plurality of pixels corresponding to the grain includes classifying each grain based on a distribution of filtered pixel values corresponding to the grain.

Assignees

• Zeiss Carl Microscopy Gmbh

Inventors

• Andrew Matthew
• Thompson William
• Correa Joaquin
• Hill Edward
• Tordoff Benjamin
• Moralee Richard

Classifications

• G06V20/698

  Matching; Classification · CPC title

• G06V10/82

  using neural networks · CPC title

• G06V30/2504

  Coarse or fine approaches, e.g. resolution of ambiguities or multiscale approaches · CPC title

• G06F18/214

  Generating training patterns; Bootstrap methods, e.g. bagging or boosting · CPC title

• G06F18/23

  Clustering techniques · CPC title