Systems and methods for improved core sample analysis

1. A method comprising:

receiving, by a computing device, a plurality of core images for a plurality of core samples, wherein each core image of the plurality of core images comprises at least one image of a core sample of the plurality of core samples;

determining, by a machine learning model, based on each core image of the plurality of core images, an orientation line for each core sample of the plurality of core samples;

generating, by the machine learning model, based on the orientation line associated with each core sample of the plurality of core samples, one or more ellipses associated with one or more points of each core sample of the plurality of core samples;

determining, based on the one or more ellipses associated with the one or more points of each core sample of the plurality of core samples, one or more attributes of each core sample of the plurality of core samples; and

providing, at a user interface, for a first core image of the plurality of core images, an output image comprising the at least one image of the core sample and an overlay of a segmentation mask, wherein the segmentation mask is indicative of the one or more attributes associated with the first core image.

2. The method of claim 1 , wherein the machine learning model comprises one or more of: an inference model, a segmentation model, or a convolutional neural network.

3. The method of claim 1 , wherein the one or more attributes comprise one or more core edges, one or more fractures, one or more broken zones, one or more beddings, or one or more veins.

4. The method of claim 1 , further comprising:

generating, based on the one or more attributes associated with the first core image, the segmentation mask.

5. The method of claim 1 further comprising:

receiving, via the user interface, an indication of one or more user edits associated with the output image; and

training, based on the indication of the one or more user edits, the machine learning model.

6. The method of claim 5 , wherein the one or more user edits comprise an adjustment of: at least one attribute of the one or more attributes, at least one adjustment of the overlay of the segmentation mask, at least one user-defined attribute associated with the at least one image of the core sample, or at least one user-defined attribute associated with the segmentation mask.

7. The method of claim 1 , wherein determining the one or more attributes of each core sample of the plurality of core samples comprises comprises:

determining, based on an orientation of the one or more ellipses in relation to the orientation line, a beta angle associated with the one or more points of each core sample of the plurality of core samples;

determining, based on the one or more ellipses and an axis of each core sample of the plurality of core samples, an alpha angle associated with the one or more points of each core sample of the plurality of core samples; and

determining, based on each alpha angle and each beta angle, the one or more attributes of each core sample of the plurality of core samples.

8. A method comprising:

determining, by a machine learning model, based on at least one image of a core sample, an orientation line for the core sample;

determining, by the machine learning model, based on one or more ellipses associated with one or more points of the core sample, one or more attributes of the core sample, wherein the one or more ellipses are based on the orientation line and the one or more points of the core sample;

generating, by the machine learning model, a segmentation mask indicative of the one or more attributes;

providing, via a user interface, an output image comprising the at least one image of the core sample and an overlay of the segmentation mask;

receiving, via the user interface, an indication of one or more user edits associated with the output image; and

training, based on the indication of the one or more user edits, the machine learning model.

9. The method of claim 8 , wherein training the machine learning model comprises:

receiving expert annotation associated with the one or more user edits; and

training, based on the expert annotation, the machine learning model.

10. The method of claim 9 , wherein the expert annotation comprises an indication of an acceptance of the one or more user edits, a rejection of the one or more user edits, or an adjustment to the one or more user edits.

11. The method of claim 8 , wherein the one or more attributes comprise one or more core edges, one or more fractures, one or more broken zones, one or more beddings, or one or more veins.

12. The method of claim 8 , wherein the one or more user edits comprise: an adjustment of at least one attribute of the one or more attributes, at least one adjustment of the overlay of the segmentation mask, at least one user-defined attribute associated with the at least one image of the core sample, or at least one user-defined attribute associated with the segmentation mask.

13. The method of claim 8 , wherein the machine learning model comprises a trained segmentation model, and wherein training, based on the indication of the one or more user edits, the machine learning model comprises: re-training, based on the indication of the one or more user edits, the trained segmentation model.

14. The method of claim 13 , further comprising:

generating, based on the re-trained segmentation model, a second segmentation mask indicative of:

the one or more attributes associated with the at least one image of the core sample, and

the one or more user edits;

and

providing, via the user interface, a second output image comprising the at least one image of the core sample and an overlay of the second segmentation mask.

15. A system comprising:

a computing device configured to receive at least one image of a core sample;

a machine learning module configured to:

determine, based on at least one image of a core sample, an orientation line for the core sample;

determine, based on one or more ellipses associated with one or more points of the core sample, one or more attributes of the core sample, wherein the one or more ellipses are based on the orientation line and the one or more points of the core sample; and

generate a segmentation mask indicative of the one or more attributes;

a user interface configured to:

provide an output image comprising the at least one image of the core sample and an overlay of the segmentation mask; and

receive an indication of one or more user edits associated with the output image; and

the machine learning module further configured to train, based on the indication of the one or more user edits, the machine learning model.

16. The system of claim 15 , wherein the machine learning module is further configured to:

receive expert annotation associated with the one or more user edits; and

retrain, based on the expert annotation, the machine learning model.

17. The system of claim 16 , wherein the expert annotation comprises an indication of an acceptance of the one or more user edits, a rejection of the one or more user edits, or an adjustment to the one or more user edits.

18. The system of claim 15 , wherein the one or more attributes comprise one or more core edges, one or more fractures, one or more broken zones, one or more beddings, or one or more veins.

19. The system of claim 15 , wherein the one or more user edits comprise: an adjustment of at least one attribute of the one or more attributes, at least one adjustment of the overlay of the segmentation mask, at least one user-defined attribute associated with the at least one image of the core sample, or at least one user-defined attribute associated with the segmentation mask.

20. The system of claim 15 , wherein the machine learning model comprises a trained segmentation model, and wherein the machine learning module is further configured to train, based on the indication of the one or more user edits, the machine learning model by: re-training, based on the indication of the one or more user edits, the trained segmentation model.