Systems and methods for visual verification

Patent metadata

Field	Value
Publication number	US-10475038-B1
Application number	US-201816199302-A
Country	US
Kind code	B1
Filing date	Nov 26, 2018
Priority date	Nov 26, 2018
Publication date	Nov 12, 2019
Grant date	Nov 12, 2019

Abstract

A system including: a document holder configured to receive a document; one or more optical sensors directed toward the document holder; one or more processors; and at least one memory having stored thereon computer program code that, when executed by the processor, instructs the processor to: control the one or more optical sensors to capture a plurality of images of the document within the document holder; identify a feature of the document within each of the plurality of images, the feature having visual characteristics that differ based on an angle at which the feature is viewed; and output, in response comparing the respective visual characteristics of the feature of the document in each of the plurality of images to corresponding expected visual characteristics of the feature for the document, an indication of a verification of the document.

First claim

We claim:

1. A system comprising:
a document holder configured to receive a document;
one or more cameras directed toward the document holder;
one or more processors; and
at least one memory having stored thereon computer program code that, when executed by the processor, instructs the processor to:
control the one or more cameras to capture a plurality of images of the document within the document holder;
identify a feature of the document within each of the plurality of images, the feature having visual characteristics that differ based on an angle at which the feature is viewed;
determine, based on an identity of the document, expected visual characteristics of the feature for the document, the expected visual characteristics differing based on an angle at which the feature is viewed;
compare, through image analysis, the respective visual characteristics of the feature of the document in each of the plurality of images to the expected visual characteristics of the feature for the document; and
output, in response to the comparing, an indication of a verification of the document.

2. The system of claim 1 comprising a plurality of cameras disposed about the document holder.

3. The system of claim 1 further comprising a motor configured to rotate the document holder,
wherein the computer program code further instructs the processor to control the motor to rotate the document holder while controlling the one or more cameras to capture the plurality of images.

4. The system of claim 1 further comprising a motor configured to move the one or more cameras about the document holder,
wherein the computer program code further instructs the processor to control the motor to move the one or more cameras while controlling the one or more cameras to capture the plurality of images.

5. The system of claim 1 , wherein
the computer program code further instructs the processor to estimate, for each of the plurality of images and based on a relative positioning of the document holder to the one or more-cameras, a respective viewing angle of the document within each of the plurality of images, and
the comparing comprises comparing, for each of the plurality of images, the respective visual characteristics of the feature of the document to the corresponding expected visual characteristics of the expected visual characteristics at the respective estimated viewing angle of the document.

6. The system of claim 1 , wherein
the plurality of images comprises a sequence of images, and
the comparing further comprises matching a sequence of the respective visual characteristics of the feature of the document within the sequence of images to an expected sequence of the expected visual characteristics for the sequence of image.

7. A method comprising:
receiving a plurality of images of a document, each of the plurality of images having a respective perspective of the document;
identifying a feature of the document visible within each of the plurality of images, the feature having visual characteristics that differ based on a perspective at which the feature is viewed;
determining, based on an identity of the document, expected visual characteristics of the feature for the document, the expected visual characteristics differing based on a perspective at which the feature is viewed;
comparing, through image analysis, the respective visual characteristics of the feature of the document in each of the plurality of images to the expected visual characteristics of the feature for the document; and
outputting, in response to the comparing, an indication of verification of the document.

8. The method of claim 7 , wherein
the feature comprises one or more lenticular lens, and
a geometry of the one or more lenticular lenses is assigned to document.

9. The method of claim 8 further comprising:
receiving an indication of the identity of the document; and
determining an expected geometry of the lenticular lens based on the indicated identity of the document,
wherein the comparing is based on the expected geometry of the lenticular lens.

10. The method of claim 7 , wherein
the feature comprises one or more lenticular lenses,
at least one of a shape and a number of lenticular lenses is assigned to the document,
the method further comprises:
receiving an indication of the identity of the document; and
determining at least one of an expected shape or an expected number of the one or more lenticular lenses based on the indicated identity, and

the comparing is based on the at least one expected shape or expected number of the one or more lenticular lenses.

11. The method of claim 7 , wherein
the visual characteristics of the feature are assigned to the document, and
the method further comprises:
receiving an indication of the identity of the document; and
determining the expected visual characteristics for the feature based on the indicated identity.

12. The method of claim 7 further comprising estimating, for each of the plurality of images, a respective viewing angle of the document within each of the plurality of images,
wherein the comparing comprises comparing, for each of the plurality of images, the visual characteristics of the feature to the corresponding expected visual characteristics of the expected visual characteristics at respective estimated viewing angle of the document.

13. The method of claim 7 , wherein the feature comprises a hologram or a physical three-dimensional feature.

14. The method of claim 7 , wherein the plurality of images comprises a video.

15. The method of claim 14 , wherein the comparing comprises comparing an order of the visual characteristics of the feature in the video to an expected order based on a relative motion of the document within the video.

16. The method of claim 7 , wherein the document comprises a credit card.

17. The method of claim 7 , wherein the visual characteristics of the feature comprise a machine-readable code.

18. The method of claim 7 , wherein
the document comprises a plurality of features, respective visual characteristics of each of the plurality of features being different based on a perspective at which the respective feature is viewed, and
the method comprises:
identifying each of the plurality of features of the document within the plurality of images;
determining, based on the identity of the document, respective expected visual characteristics of each of the plurality of features;
comparing the respective visual characteristics of the respective visual characteristics of each of the features to corresponding to the respective expected visual characteristics of the feature for the document; and
outputting, in response to the comparing, the indication of verification of the document.

19. The method of claim 18 , further comprising estimating, for each of the plurality of images, a respective viewing angle of the document within each of the plurality of images,
wherein the comparing comprises comparing, for each of the plurality of images, the respective visual characteristics of each of the respective features to the corresponding expected visual characteristics at the respective estimated viewing angle.

Assignees

• Capital One Services Llc

Inventors

• Osborn Kevin
• Marsch Daniel

Classifications

• G06Q20/401

  Transaction verification · CPC title

• G06Q20/042

  characterized in that the payment protocol involves at least one cheque · CPC title

• G07D7/2016 Primary

  using feature extraction, e.g. segmentation, edge detection or Hough-transformation · CPC title

• G06V30/40

  Document-oriented image-based pattern recognition · CPC title

• G06V10/147

  Details of sensors, e.g. sensor lenses (fingerprint or palmprint sensors G06V40/13; vascular sensors G06V40/145; eye sensors G06V40/19) · CPC title