System and methods for improved sample imaging
AssigneeVeracio Ltd.
InventorsLuke George, Sasa Krneta
Life of this patent
Estimating the term…
Abstract
Provided herein are methods and systems for improved sample imaging. An input image of a sample may be aligned with at least one hyperspectral image of a portion of the sample. Once aligned, the input image and the at least one hyperspectral image may be provided to at least one machine learning model to generate a calibration matrix. The calibration matrix may be applied to the input image to generate a false color mineral map. The false color mineral map may be indicative of a plurality of minerals associated with the sample. Additionally, the false color mineral map may comprise a size that is much smaller than a false color mineral map generated using a hyperspectral image(s) of the entire sample.
Claims20
A method comprising:
- receiving, by a computing device, an input image and at least one hyperspectral image, wherein the input image depicts a sample comprising a plurality of minerals, and wherein the at least one hyperspectral image depicts a portion of the sample;
- determining, by at least one machine learning model, an alignment of the input image and the at least one hyperspectral image;
- generating, by the at least one machine learning model, based on the alignment of the input image and the at least one hyperspectral image, a calibration matrix;
- and
- generating, based on the calibration matrix and the input image, a false color mineral map indicative of the plurality of minerals associated with the sample.
A method comprising:
- receiving, by a computing device, a plurality of training input images and a plurality of training hyperspectral images;
- receiving a plurality of false color mineral maps and a plurality of calibration matrices associated with the plurality of training samples;
- and
- training at least one machine learning model based on:
- the plurality of training input images, the plurality of training hyperspectral images, the plurality of false color mineral maps, and the plurality of calibration matrices, wherein the at least one machine learning model, once trained, is configured to generate a calibration matrix for an input image of a sample, and wherein the calibration matrix is based on at least one hyperspectral image depicting a portion of the sample.
An apparatus comprising:
- one or more processors;
- and
- computer-executable instructions that, when executed by the one or more processors, cause the apparatus to:
- receive an input image and at least one hyperspectral image, wherein the input image depicts a sample comprising a plurality of minerals, and wherein the at least one hyperspectral image depicts a portion of the sample;
- determine, via at least one machine learning model, an alignment of the input image and the at least one hyperspectral image;
- generate, via the at least one machine learning model, based on the alignment of the input image and the at least one hyperspectral image, a calibration matrix;
- and
- generate, based on the calibration matrix and the input image, a false color mineral map indicative of the plurality of minerals associated with the sample.
Dependencies are inferred from the claim text.