Systems and methods for synthesizing a diamond using machine learning

Patent metadata

Field	Value
Publication number	US-11698347-B2
Application number	US-202117327456-A
Country	US
Kind code	B2
Filing date	May 21, 2021
Priority date	May 22, 2020
Publication date	Jul 11, 2023
Grant date	Jul 11, 2023

Abstract

Disclosed herein are systems and methods for synthesizing a diamond using a diamond synthesis machine. A processor receives a plurality of images of a diamond during synthesis within a diamond synthesis machine, each of the plurality of images captured within a time period. The processor executes a diamond state prediction machine learning model using the plurality of images to obtain a predicted data object, the predicted data object indicating a predicted state of the diamond within the diamond synthesis machine at a time subsequent to the time period. The processor detects a predicted defect, a number of defects, defect types, and/or sub-features of such defects and/or other characteristics (e.g., a predicted shape, size, and/or other properties of predicted contours for the diamond and/or pocket holder) of the predicted state of the diamond. The processor adjusts operation of the diamond synthesis machine.

First claim

What we claim is:

1. A method for synthesizing a diamond using a diamond synthesis machine, comprising:
receiving, by a processor, a plurality of images of a diamond during synthesis within a diamond synthesis machine, each of the plurality of images captured within a time period;
executing, by the processor, a diamond state prediction machine learning model using the plurality of images to obtain a predicted data object, the predicted data object indicating a predicted state of the diamond within the diamond synthesis machine at a time subsequent to the time period;
detecting, by the processor, a predicted defect in the diamond based on the predicted state of the diamond; and
adjusting, by the processor, operation of the diamond synthesis machine responsive to detecting the defect.

2. The method of claim 1 , wherein the predicted data object comprises a predicted image depicting the diamond in the predicted state, and wherein the predicted image comprises a plurality of pixels, further comprising:
executing, by the processor, a feature extraction machine learning model to extract a plurality of classification labels for the plurality of pixels of the predicted image;
wherein detecting the predicted defect in predicted state of the diamond comprises detecting the predicted defect based on the plurality of classification labels.

3. The method of claim 2 , wherein executing the feature extraction machine learning model to extract the plurality of classification labels comprises executing the defect detection machine learning model to extract a diamond top label, a diamond side label, a diamond holder label, a macroscopic defect label, a microscopic defect label, a center defect label, or an edge defect label for the plurality of pixels of the predicted image.

4. The method of claim 1 , wherein executing the diamond state prediction machine learning model comprises executing, by the processor, the diamond state prediction machine learning model using the plurality of images and a schedule of operating parameters of the diamond synthesis machine to obtain the predicted data object, the schedule of operating parameters of the diamond synthesis machine corresponding to times in which the plurality of images were captured.

5. The method of claim 4 , wherein the schedule of operating parameters further comprises a first set of operating parameters for the diamond synthesis machine for a first time between the time period and the time subsequent to the time period.

6. The method of claim 1 , wherein executing the diamond state prediction machine learning model comprises:
causing, by the processor, the diamond state prediction machine learning model to:
convert the plurality of images into a multi-dimensional vector corresponding to a predicted first state of the diamond at a first time-step subsequent to the time period; and
generate the predicted data object based on the multi-dimensional vector.

7. The method of claim 1 , wherein the predicted data object comprises a predicted image depicting the diamond in the predicted state, and wherein detecting the predicted defect comprises:
evaluating, by the processor, a shape of the diamond depicted in the predicted image according to a set of diamond defect rules; and
detecting, by the processor, a macroscopic defect, microscopic defect, a center defect, or an edge defect in the diamond responsive to determining the shape of the diamond satisfies a diamond defect rule.

8. The method of claim 1 , further comprising:
receiving, by the processor, a second plurality of images of a second diamond during synthesis within the diamond synthesis machine, each of the second plurality of images captured within a second time period;
executing, by the processor, the diamond state prediction machine learning model using the second plurality of images to obtain a second predicted data object, the second predicted data object indicating a second predicted state of the second diamond within the diamond synthesis machine at a second time subsequent to the second time period;
determining, by the processor, there are not any defects in the diamond in the predicted state; and
generating, by the processor, a record indicating no defects were detected in the diamond in the predicted state.

9. The method of claim 1 , wherein adjusting operation of the diamond synthesis machine comprises adjusting, by the processor, a pressure or a temperature of the diamond synthesis machine.

10. The method of claim 1 , wherein the predicted data object comprises a predicted image depicting the diamond in the predicted state, and wherein detecting the predicted defect in the diamond depicted in the predicted image comprises:
executing, by the processor, a feature extraction machine learning model using the predicted image to obtain a contour of the predicted defect in the diamond;
identifying, by the processor, the contour of the predicted defect; and
detecting, by the processor, the predicted defect in response to identifying the contour.

11. The method of claim 1 , wherein the predicted data object comprises a predicted image depicting the diamond in the predicted state, and wherein detecting the predicted defect in the diamond depicted in the predicted image comprises:
executing, by the processor, a feature extraction machine learning model using the predicted image to obtain a contour of the diamond or a diamond holder;
identifying, by the processor, a size or shape of the diamond based on the contour; and
detecting, by the processor, the predicted defect in response to identifying the size or shape.

12. The method of claim 1 , further comprising:
receiving, by the processor, a second plurality of images of a second diamond during synthesis within the diamond synthesis machine, each of the second plurality of images captured within a second time period;
executing, by the processor, the diamond state prediction machine learning model using the second plurality of images to obtain a second predicted data object, the second predicted data object indicating a second state of the second diamond within the diamond synthesis machine at a second time subsequent to the second time period;
determining, by the processor, a difference between the second predicted data object and an expected data object; and
training, by the processor, the diamond state prediction machine learning model according to the determined difference.

13. The method of claim 1 , wherein the predicted data object comprises a value, a vector, or an image.

14. A system for synthesizing a diamond using a diamond synthesis machine, the system comprising:
a processor configured to execute instructions stored on a non-transitory computer-readable medium to:
receive a plurality of images of a diamond during synthesis within a diamond synthesis machine, each of the plurality of images captured within a time period;
execute a diamond state prediction machine learning model using the plurality of images to obtain a predicted data object, the predicted data object indicating a predicted state of the diamond within the diamond synthesis machine at a time subsequent to the time period;
detect a predicted defect in the diamond based on the predicted state of the diamond; and
adjust operation of the diamond synthesis machine responsive to detecting the defect.

15. The system of claim 14 , wherein the predicted data object comprises a predicted image depicting the diamond in the predicted state, wherein the predicted image comprises a plurality of pixels, and wherein the processor is further configured to:
execute a defect detectio

Assignees

• Fraunhofer Usa Inc
• Univ Maryland

Inventors

• Mekala Rohan Reddy
• Muehle Matthias
• Porter Adam
• Lindvall Mikael
• Becker Michael

Classifications

• G01N21/8851 Primary

  Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges (G01N21/8806 and G01N21/93 - G01N21/95692 take precedence; optical measurement of dimensions G01B11/00; optical scanning G02B26/10; image transformation G06T3/00; computerised image enhancement G06T5/00; image processing per se for flaw detection G06T7/0002) · CPC title

• G01N21/95607

  using a comparative method · CPC title

• G01N2021/95615

  with stored comparision signal · CPC title

• G01N2021/8854

  Grading and classifying of flaws · CPC title

• G01N2021/8887

  based on image processing techniques · CPC title