Propensity model based optimization

Patent metadata

Field	Value
Publication number	US-10817997-B2
Application number	US-201816128401-A
Country	US
Kind code	B2
Filing date	Sep 11, 2018
Priority date	Sep 11, 2017
Publication date	Oct 27, 2020
Grant date	Oct 27, 2020

Abstract

Apparatuses, systems, methods, and computer program products are presented for a propensity module based optimization. A mobile computing device comprises one or more of a camera, a network interface, and an optimization module. An optimization module is configured to process one or more images from a camera using machine learning to determine a likelihood that the one or more images will pass submission to a pass/fail interface over a network interface. An optimization module is configured to submit one or more images to a pass/fail interface over a network interface in response to a likelihood satisfying a threshold.

First claim

What is claimed is:

1. An apparatus, comprising:
a mobile computing device comprising:
a camera;
a network interface; and
an optimization module configured to:
process one or more images from the camera using a machine learning propensity model that is trained on previous submissions to a pass/fail interface of a third party service provider that is remote to the mobile computing device;
determine a likelihood that the one or more images will pass submission to the pass/fail interface over the network interface by inputting the one or more images into the propensity model prior to submitting the electronic submission to the pass/fail interface of the third party service provider; and
submit the one or more images to the pass/fail interface over the network interface in response to the likelihood satisfying a threshold.

2. The apparatus of claim 1 , wherein the optimization module is further configured to:
perform a corrective action for the one or more images in response to the likelihood failing to satisfy the threshold;
redetermine the likelihood that the one or more images will pass submission to the pass/fail interface in response to performing the corrective action; and
submit the one or more images to the pass/fail interface over the network interface in response to the redetermined likelihood satisfying the threshold.

3. The apparatus of claim 2 , wherein the corrective action comprises suggesting, on the mobile computing device, an adjustment to the one or more images to a user of the mobile computing device.

4. The apparatus of claim 3 , wherein the suggestion comprises one or more of:
retaking at least one of the one or more images with a different focal length;
retaking at least one of the one or more images with a different background;
retaking at least one of the one or more images with a different field of capture;
rotating at least one of the one or more images;
swapping two or more of the one or more images;
providing additional information regarding the one or more images; and
adding a signature to an object of at least one of the one or more images and retaking the at least one of the one or more images.

5. The apparatus of claim 2 , wherein the corrective action comprises an automatic action correcting one or more characteristics of the one or more images.

6. The apparatus of claim 5 , wherein the automatic action comprises one or more of rotating at least one of the one or more images and swapping two or more of the one or more images.

7. The apparatus of claim 1 , wherein the one or more images are of a check and the pass/fail interface comprises a remote deposit interface for depositing the check using the network interface.

8. The apparatus of claim 7 , wherein the likelihood fails to satisfy the threshold in response to one or more of:
the check comprising a money order;
the one or more images of the check satisfying a blurriness threshold;
a background of the one or more images satisfying a complexity threshold;
a background of the one or more images satisfying a similarity threshold relative to the check;
the one or more images having a wrong rotation;
a front image of the check and a back image of the check being switched in the one or more images;
a signature endorsement missing from the check; and
a monetary amount written on the check failing to match a monetary amount input by a user of the mobile computing device.

9. The apparatus of claim 7 , wherein the optimization module is further configured to suggest a monetary amount for the check to a user of the mobile computing device for confirmation, the optimization module configured to determine the suggested monetary amount based on optical character recognition of a handwritten amount from the one or more images of the check.

10. The apparatus of claim 1 , wherein the optimization module is configured to determine the likelihood using a neural network based propensity model trained on previous submissions to the pass/fail interface.

11. A method, comprising:
receiving an electronic submission for a pass/fail interface of a third party service provider;
mimicking the pass/fail interface on a device that is remote from the third party service provider using a machine learning propensity model that is trained on previous submissions to the pass/fail interface of the third party service provider;
determining a likelihood that the electronic submission will be accepted by the pass/fail interface by inputting the electronic submission into the propensity model prior to submitting the electronic submission to the pass/fail interface of the third party service provider;
submitting the electronic submission to the pass/fail interface in response to the likelihood satisfying a threshold; and
performing a corrective action for the electronic submission in response to the likelihood failing to satisfy the threshold.

12. The method of claim 11 , further comprising:
redetermining the likelihood that the electronic submission will be accepted by the pass/fail interface in response to performing the corrective action;
submitting the electronic submission to the pass/fail interface in response to the redetermined likelihood satisfying the threshold; and
performing an additional corrective action for the electronic submission in response to the redetermined likelihood failing to satisfy the threshold.

13. The method of claim 11 , wherein the corrective action comprises suggesting an adjustment to the electronic submission to a user associated with the electronic submission.

14. The method of claim 13 , wherein the suggestion comprises one or more of:
retaking a photo for the electronic submission with a different focal length;
retaking a photo for the electronic submission with a different background;
retaking a photo for the electronic submission with a different field of capture;
rotating a photo for the electronic submission;
swapping photos of the electronic submission;
adding information to the electronic submission; and
adding a signature to the electronic submission.

15. The method of claim 11 , wherein the corrective action comprises an automatic action correcting one or more characteristics of the electronic submission.

16. The method of claim 15 , wherein the automatic action comprises one or more of rotating a photo of the electronic submission and swapping front and back photos of the electronic submission.

17. The method of claim 11 , wherein the electronic submission comprises one or more photos of a check and the pass/fail interface comprises a remote deposit interface for depositing the check.

18. The method of claim 17 , wherein the likelihood fails to satisfy the threshold in response to one or more of:
the check comprising a money order;
the one or more photos of the check satisfying a blurriness threshold;
a background of the one or more photos satisfying a complexity threshold;
a background of the one or more photos satisfying a similarity threshold relative to the check;
the one or more photos having a wrong rotation;
a front photo of the check and a back photo of the check being switched in the one or more photos;
a signature endorsement missing from the check; and
a monetary amount written on the check failing to match a monetary amount input by a user.

19. The method of claim 11 , wherein the electronic submission comprises a database query and the pass/fail interface comprises a database interface.

20. An apparatus, comprising:
means for determining a

Assignees

• Mx Tech Inc

Inventors

• Dewitt Brandon
• Mcbride Ryan
• Smit Shane
• Bodily Josh

Classifications

• G06V30/146 Primary

  Aligning or centring of the image pick-up or image-field · CPC title

• G06T7/0002 Primary

  Inspection of images, e.g. flaw detection · CPC title

• G06V30/10

  Character recognition · CPC title

• G06V30/416

  Extracting the logical structure, e.g. chapters, sections or page numbers; Identifying elements of the document, e.g. authors · CPC title

• G06V10/24

  Aligning, centring, orientation detection or correction of the image · CPC title