Generation of digital well schematics

1. A computing device comprising:

a processing device; and

a non-transitory memory device comprising instructions that are executable by the processing device to cause the processing device to perform operations comprising:

obtaining an image of a physical media wellbore schematic diagram for a wellbore;

identifying a layout type of the physical media wellbore schematic diagram in the image;

performing text recognition and image recognition on the image to detect text information and object information based on the identified layout type;

classifying the text information into content features using a first classification model;

classifying the object information into component features using a second classification model;

editing the image to remove the classified component features;

performing image recognition on the edited image to detect additional object information;

classifying the additional object information into additional component features using a third classification model; and

applying engineering constraints to the content features and component features to produce structured wellbore schematic data.

2. The computing device of claim 1 , the operations further comprising:

applying a generation model to the structured wellbore schematic data to generate a digital wellbore schematic corresponding to the physical media wellbore schematic diagram.

3. The computing device of claim 2 wherein applying a generation model to the structured wellbore schematic data to generate a digital wellbore schematic corresponding to the physical media wellbore schematic diagram comprises:

extracting component features and content features from the structured wellbore schematic data; and

generating a digital wellbore schematic diagram including visual representation of the content features and component features.

4. The computing device of claim 1 , wherein the additional object information relates to additional component features that differ in at least one of size or type from the component features related to the object information.

5. The computing device of claim 1 , the operations further comprising:

comparing the structured wellbore schematic data to a second structured wellbore schematic data associated with a second wellbore to identify differences in component features and content features; and

adjusting drilling operations associated with at least one of the wellbore or the second wellbore, based on the identified differences in component features or content features.

6. The computing device of claim 1 , wherein component features include at least two of casing, tubing, perforation, liner, packer, or machinery information and wherein content features include at least two of depth information, location information, assess procurement, length of time drilled, or remaining asset.

7. A non-transitory computer-readable medium that includes instructions that are executable by a processing device for causing the processing device to perform operations comprising:

obtaining an image of a physical media wellbore schematic diagram for a wellbore;

identifying a layout type of the physical media wellbore schematic diagram in the image;

performing text recognition and image recognition on the image to detect text information and object information based on the identified layout type;

classifying the text information into content features using a first classification model,. wherein the content features include at least two of depth information, location information, assess procurement, length of time drilled, or remaining asset;

classifying the object information into component features using a second classification model, wherein the component features include at least two of casing, tubing, perforation, liner, packer, or machinery information; and

applying engineering constraints to the content features and component features to produce structured wellbore schematic data.

8. The non-transitory computer-readable medium of claim 7 , wherein the operations further comprise applying a generation model to the structured wellbore schematic data to generate a digital wellbore schematic corresponding to the physical media wellbore schematic diagram.

9. The non-transitory computer-readable medium of claim 8 , wherein applying a generation model to the structured wellbore schematic data to generate a digital wellbore schematic corresponding to the physical media wellbore schematic diagram comprises:

extracting component features and content features from the structured wellbore schematic data; and

generating a digital wellbore schematic diagram including visual representation of the content features and component features.

10. The non-transitory computer-readable medium of claim 7 , wherein the operations further comprise:

editing the image to remove the classified component features;

performing image recognition on the edited image to detect additional object information; and

classifying the additional object information into additional component features using a third classification model.

11. The non-transitory computer-readable medium of claim 10 , wherein the additional object information relates to the additional component features that differ in at least of size or type from the component features related to the object information.

12. The non-transitory computer-readable medium of claim 7 , wherein the operations further comprise:

comparing the structured wellbore schematic data to a second structured wellbore schematic data associated with a second wellbore to identify differences in component features and content features; and

adjusting drilling operations associated with at least one of the wellbore or the second wellbore, based on the identified differences in component features or content features.

13. A method comprising:

obtaining an image of a physical media wellbore schematic diagram for a wellbore;

identifying a layout type of the physical media wellbore schematic diagram in the image;

performing text recognition and image recognition on the image to detect text information and object information based on the identified layout type;

classifying the text information into content features using a first classification model;

classifying the object information into component features using a second classification model; and

applying engineering constraints to the content features and component features to produce structured wellbore schematic data;

comparing the structured wellbore schematic data to a second structured wellbore schematic data associated with a second wellbore to identify differences in component features and content features; and

adjusting drilling operations associated with at least one of the wellbore or the second wellbore, based on the identified differences in component features or content features.

14. The method of claim 13 , further comprising:

applying a generation model to the structured wellbore schematic data to generate a digital wellbore schematic corresponding to the physical media wellbore schematic diagram.

15. The method of claim 14 wherein applying a generation model to the structured wellbore schematic data to generate a digital wellbore schematic corresponding to the physical media wellbore schematic diagram comprises:

extracting component features and content features from the structured wellbore schematic data; and

generating a digital wellbore schematic diagram including visual representation of the content features and component features.

16. The method of claim 13 , further comprising:

editing the image to remove the classified component features;

performing image recognition on the edited image to detect additional object information; and

classifying the additional object information into additional component features using a third classification model.

17. The method of claim 16 , wherein the additional object information relates to additional component features that differ in at least one of size or type from the component features related to the object information.

18. The method of claim 13 , wherein component features include at least two of casing, tubing, perforation, liner, packer, or machinery information, and wherein content features include at least two of depth information, location information, assess procurement, length of time drilled, or remaining asset.

19. The method of claim 13 , further comprising:

pre-processing the physical media wellbore schematic diagram to generate a pre-processed schematic; and

applying the first classification model to the pre-processed schematic to detect the component features.

20. The method of claim 13 , wherein the engineering constraints include rules for well component placement, kick-off point (KoP) placement, asset extraction strategies, extraction exclusion zones, or depth restrictions.