Securing user-entered text in-transit

1. A computer-implemented method comprising:

identifying, based on an automated analysis of a document, two or more fields of user-entered text;

extracting each of the two or more fields of user-entered text from the document;

tagging each of the two or more extracted fields of user-entered text with a field identifier and a location of the user-entered text within the document;

converting a first field of user-entered text, of the two or more fields of extracted and tagged user-entered text, into a first image using a neural network, wherein the first image comprises:

a graphical representation of the first field of user-entered text that is configured to obfuscate the first field of user-entered text while the document is in-transit to a second device; and

a first tag, indicating a first field and a first text location, that is used to recreate the document; and

transmitting, to the second device, the first image and the document.

2. The computer-implemented method of claim 1 , comprising:

encrypting, prior to the transmitting the first image and the document, the first image using an encryption key and an encryption algorithm; and

transmitting, to the second device, the encrypted first image and the document.

3. The computer-implemented method of claim 1 , wherein the automated analysis of the document comprises image segmentation to identify the two or more fields of user-entered text.

4. The computer-implemented method of claim 1 , wherein the automated analysis of the document comprises background subtraction to identify the two or more fields of user-entered text.

5. The computer-implemented method of claim 1 , wherein converting the first field of user-entered text into the first image comprises an image-to-image transformation.

6. The computer-implemented method of claim 1 , wherein converting the first field of user-entered text into the first image comprises a text-to-image transformation.

7. The computer-implemented method of claim 1 , wherein converting the first field of user-entered text into the first image comprises a font transformation.

8. A computing device comprising:

one or more processors; and

memory storing instructions that, when executed by the one or more processors, cause the computing device to:

identify two or more fields of user-entered text in a document using an automated analysis;

extract each of the two or more fields of user-entered text from the document;

tag each of the two or more extracted fields of user-entered text with a field identifier and a location of the user-entered text within the document;

convert a first field of user-entered text, of the two or more fields of extracted and tagged user-entered text, into a first image using a neural network, wherein the first image comprises:

a graphical representation of the first field of user-entered text that is configured to obfuscate the first field of user-entered text while the document is in-transit to a second device; and

a first tag, indicating a first field and a first text location, that is used to recreate the document; and

transmit the first image and the document to the second device.

9. The computing device of claim 8 , wherein the instructions, when executed by the one or more processors, cause the computing device to:

encrypt, prior to the transmitting the first image and the document, the first image using an encryption key and an encryption algorithm; and

transmit the encrypted first image and the document to the second device.

10. The computing device of claim 8 , wherein the automated analysis of the document comprises at least one of:

image segmentation to identify the two or more fields of user-entered text; or

background subtraction to identify the two or more fields of user-entered text.

11. The computing device of claim 8 , wherein converting the first field of user-entered text into the first image comprises at least one of:

an image-to-image transformation;

a text-to-image transformation; or

a font transformation.

12. A computer-implemented method comprising:

receiving, by a second device and from a first device, a first image and a document, wherein the first image comprises a first tag indicating a first field and first text location;

reconstructing, using a neural network, a first field of user-entered text from the first image;

creating a complete document by combining the first field of user-entered text and the received document, using the first tag;

comparing the first field of user-entered text to previously registered user data to determine whether the first field of user-entered text and the previously registered user data match; and

verifying, based on a determination that the first field of user-entered text matches the previously registered user data, the complete document.

13. The computer-implemented method of claim 12 , wherein the first image comprises an encrypted image.

14. The computer-implemented method of claim 13 , comprising:

decrypting the encrypted image prior to reconstructing the first field of user-entered text.

15. The computer-implemented method of claim 12 , wherein reconstructing the first field of user-entered text from the first image comprises at least one of:

an image-to-image transformation;

an image-to-text transformation; or

a font transformation.

16. A computing device comprising:

one or more processors; and

memory storing instructions that, when executed by the one or more processors, cause the computing device to:

receive a first image and a document, wherein the first image comprises a tag indicating a first field and first text location;

reconstruct a first field of user-entered text from the first image using a neural network;

create a complete document by combining the first field of user-entered text and the received document, using the tag;

compare the first field of user-entered text to previously registered user data to determine whether the first field of user-entered text and the previously registered user data match; and

verify, based on a determination that the first field of user-entered text matches the previously registered user data, the complete document.

17. The computing device of claim 16 , wherein the first image comprises a first encrypted image.

18. The computing device of claim 17 , wherein the instructions, when executed by the one or more processors, cause the computing device to:

decrypt the first encrypted image prior to reconstructing the first field of user-entered text.

19. The computing device of claim 16 , wherein reconstructing the first field of user-entered text from the first image comprises at least one of:

an image-to-image transformation;

an image-to-text transformation; or

a font transformation.

20. The computing device of claim 16 , wherein the neural network comprises at least one of:

a generative adversarial network (GAN);

a consistent adversarial network (CAN);

a cyclic generative adversarial network (C-GAN);

a deep convolutional GAN (DC-GAN);

GAN interpolation (GAN-INT);

GAN-CLS; or

a cyclic-CAN (C-CAN).

21. The computing device of claim 16 , wherein the first field of user-entered text comprises handwritten information.

22. The computer-implemented method of claim 12 , comprising

receiving, by the second device and from the first device, a second image, wherein the second image comprises a second tag indicating a second field and second text location;

reconstructing, using the neural network, a second field of user-entered text from the second image, wherein creating the complete document further comprises combining the second field of user-entered text and the received document.