Extracting card data from multiple cards

Patent metadata

Field	Value
Publication number	US-9892317-B2
Application number	US-201615297127-A
Country	US
Kind code	B2
Filing date	Oct 18, 2016
Priority date	Jun 30, 2013
Publication date	Feb 13, 2018
Grant date	Feb 13, 2018

Abstract

Extracting financial card information with relaxed alignment comprises a method to receive an image of a card, determine one or more edge finder zones in locations of the image, and identify lines in the one or more edge finder zones. The method further identifies one or more quadrilaterals formed by intersections of extrapolations of the identified lines, determines an aspect ratio of the one or more quadrilateral, and compares the determined aspect ratios of the quadrilateral to an expected aspect ratio. The method then identifies a quadrilateral that matches the expected aspect ratio and performs an optical character recognition algorithm on the rectified model. A similar method is performed on multiple cards in an image. The results of the analysis of each of the cards are compared to improve accuracy of the data.

First claim

What is claimed is:

1. A computer-implemented method to extract financial card information from multiple cards, comprising:
identifying, by one or more computing devices, one or more quadrilaterals formed by a plurality of intersections of lines in an image of a plurality of cards;
determining, by the one or more computing devices, a corresponding aspect ratio for each of the quadrilaterals;
identifying, by the one or more computing devices, one or more particular quadrilaterals that match an expected aspect ratio based on a comparison of the determined aspect ratios for each of the quadrilaterals to an expected aspect ratio of a type of card represented in the image; and
performing, by the one or more computing devices, an optical character recognition algorithm on each of the identified particular quadrilaterals to extract card information from each quadrilateral.

2. The method of claim 1 , further comprising:
cropping, by the one or more computing devices, the image into a plurality of images corresponding to the identified quadrilaterals; and
wherein the performing step is performed on each cropped image to extract financial card information from each cropped image.

3. The method of claim 1 , further comprising:
modeling, by the one or more computing devices, in three dimensions one or more regions of the image encompassed by the one or more particular quadrilaterals;
rectifying, by the one or more computing devices, the three dimensional models to produce one or more rectified images of the particular quadrilaterals,
wherein the optical character recognition algorithm is performed on the rectified images of the card.

4. The method of claim 1 , further comprising:
determining, by the one or more computing devices, a location on each of the identified quadrilaterals comprising account information, the location being based at least in part on a type of card represented in each of the quadrilaterals; and
applying, by the one or more computing devices, the optical character recognition algorithm to only the determined location on each identified quadrilateral.

5. The method of claim 1 , wherein the image of the one or more identified quadrilaterals is not cropped, and the optical character recognition algorithm is performed on each of the identified quadrilaterals.

6. The method of claim 1 , further comprising:
extracting, by the one or more computing devices, a user name from each of the identified quadrilaterals;
comparing, by the one or more computing devices, the user names extracted from each of the identified quadrilaterals; and
assessing, by the one or more computing devices, the accuracy of the extracted names based at least in part on the comparison.

7. The method of claim 1 , further comprising converting, by the one or more computing devices, the image of the plurality of cards to grayscale.

8. The method of claim 1 , further comprising:
storing, by the one or more computing devices, the aspect ratios of one or more card types; and
identifying, by the one or more computing devices, the expected aspect ratio of the card types in the cropped images based on the stored aspect ratios.

9. The method of claim 1 , further comprising:
detecting, by the one or more computing devices, blur in the image; and
performing, by the one or more computing devices, a configured action to remove the blur from the image.

10. The method of claim 1 , wherein the card comprises a credit card, a debit card, an identification card, a loyalty card, or an access card.

11. The method of claim 1 , wherein the plurality of cards are financial cards.

12. A computer program product, comprising:
a non-transitory computer-readable storage device having computer-executable program instructions embodied thereon that when executed by a computer cause the computer to extract financial card information from multiple cards, the computer-executable program instructions comprising:
computer-executable program instructions to identify one or more quadrilaterals formed by a plurality of intersections of lines in an image of a plurality of cards;
computer-executable program instructions to determine a corresponding aspect ratio for each of the quadrilaterals;
computer-executable program instructions to identify one or more particular quadrilaterals that match an expected aspect ratio based on a comparison of the determined aspect ratios for each of the quadrilaterals to an expected aspect ratio of a type of card represented in the image;
computer-executable program instructions to crop the image into a plurality of images corresponding to the plurality of quadrilaterals identified as preferred card boundaries; and
computer-executable program instructions to perform an optical character recognition algorithm on each cropped image to extract card information from each cropped image.

13. The computer program product of claim 12 , the computer-executable program instructions further comprising:
computer-executable program instructions to modeling one or more regions of the image encompassed by the one or more identified quadrilaterals in three dimensions; and
computer-executable program instructions to rectify the three dimensional model to produce a rectified image of the particular quadrilaterals,
wherein the optical character recognition algorithm is performed on the rectified model of the card.

14. The computer program product of claim 12 , the computer-executable program instructions further comprising:
computer-executable program instructions to determine a location on each of the particular quadrilaterals comprising account information, the location being based at least in part on a type of cards represented in each of the quadrilaterals; and
computer-executable program instructions to perform the optical character recognition algorithm on only the determined location of each identified quadrilateral.

15. The computer program product of claim 12 , further comprising computer program instructions to convert the image to grayscale.

16. The computer program product of claim 12 , the computer-executable program instructions further comprising:
computer-executable program instructions to extract a user name from each of the identified quadrilaterals;
computer-executable program instructions to compare a user name from each of the identified quadrilaterals; and
computer-executable program instructions to assess the accuracy of the extracted names based at least in part on the comparison.

17. The computer program product of claim 12 , the computer-executable program instructions further comprising:
computer-executable program instructions to detect blur in the image; and
computer-executable program instructions to perform a configured action to remove the blur from the image.

18. The computer program product of claim 12 , wherein the card comprises a credit card, a debit card, an identification card, a loyalty card, or an access card.

19. The computer program product of claim 12 , the computer-executable program instructions further comprising:
computer-executable program instructions to crop the image into a plurality of images corresponding to the plurality of quadrilaterals identified as preferred card boundaries; and
computer-executable program instructions to perform the optical character recognition algorithm on each cropped image to extract card information from each cropped image.

20. The computer program product of claim 12 , wherein the image of the one or more identified quadrilaterals i

Assignees

• Google Inc
• Google Llc

Inventors

• Wang Xiaohang
• Huber Jeff
• Shamsi Farhan
• Okshtein Yakov
• Kumar Sanjiv
• Rowley Henry Allan
• Mitchell Marcus Quintana
• Repenning Debra Lin

Classifications

• G06V30/416

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

• G06V30/414 Primary

  Extracting the geometrical structure, e.g. layout tree; Block segmentation, e.g. bounding boxes for graphics or text · CPC title

• G06F18/22

  Matching criteria, e.g. proximity measures · CPC title

• G06V30/1478

  of characters or characters lines · CPC title

• G06K9/3283

  Physics · mapped topic