MOBILE DOCUMENT DETECTION AND ORIENTATION BASED ON REFERENCE OBJECT CHARACTERISTICS

• US 20160125613A1
• Filed: 10/29/2015
• Published: 05/05/2016
• Est. Priority Date: 10/30/2014
• Status: Active Grant

• Current Assignee: Kofax Incorporated
• Original Assignee: Kofax Incorporated
• Inventors: Shustorovich, Alexander, Thrasher, Christopher W., Ma, Jiyong, Macciola, Anthony, Amtrup, Jan W.
• Granted Patent: US 9,760,788 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06V 10/242 by image rotation, e.g. by 90 degrees
• CPC Class Codes: G06V 30/2253 Recognition of characters printed with magnetic ink G06V30/2247 takes precedence
• CPC Class Codes: G06V 30/414 Extracting the geometrical structure, e.g. layout tree; Block segmentation, e.g. bounding boxes for graphics or text

Abstract

In various embodiments, methods, systems, and computer program products for detecting, estimating, calculating, etc. characteristics of a document based on reference objects depicted on the document are disclosed. In one approach, a computer-implemented method for processing a digital image depicting a document includes analyzing the digital image to determine one or more of a presence and a location of one or more reference objects; determining one or more geometric characteristics of at least one of the reference objects; defining one or more region(s) of interest based at least in part on one or more of the determined geometric characteristics; and detecting a presence or an absence of an edge of the document within each defined region of interest. Additional embodiments leverage the type of document depicted in the image, multiple frames of image data, and/or calculate or extrapolate document edges rather than locating edges in the image.

Claims

1. A computer-implemented method for processing a digital image depicting a document, the method comprising:

analyzing the digital image to determine one or more of a presence and a location of one or more reference objects;

determining one or more geometric characteristics of at least one of the reference objects;

defining one or more region(s) of interest based at least in part on one or more of the determined geometric characteristics; and

detecting a presence or an absence of an edge of the document within each defined region of interest.

2. The method as recited in claim 1, wherein the geometric characteristics comprise one or more of:

an alignment angle of a longitudinal axis of a region encompassing the one or more reference object(s);

a height of at least one of the reference object(s);

a baseline corresponding to a plurality of the reference objects; and

an aspect ratio of one or more of the reference objects

an aspect ratio of a region depicting the one or more reference object(s).

3. The method as recited in claim 1, wherein detecting the presence of the edge of the document within each defined region of interest further comprises locating a transition from a background region of the digital image to a non-background region of the digital image within the defined region of interest.

4. The method as recited in claim 3, wherein the detecting further comprises determining a longitudinal axis of the located transition.

5. The method as recited in claim 4, wherein the detecting further comprises determining whether the longitudinal axis of the located transition corresponds to a longitudinal axis of a region encompassing the one or more reference object(s).

6. The method as recited in claim 5, wherein the longitudinal axes are determined to correspond in response to determining one or more of the following conditions are met:

the longitudinal axes are substantially parallel;

the longitudinal axes are substantially perpendicular; and

at least three points along the longitudinal axis of the located transition are each located within a threshold distance of at least three corresponding points along the longitudinal axis of the region encompassing the one or more reference object(s).

7. The method as recited in claim 1, wherein the reference object(s) comprise one or more of magnetic ink character recognition (MICR) characters and machine readable zone (MRZ) characters.

8. A computer-implemented method for processing a plurality of digital image frames, each frame comprising at least a partial digital representation of a document, and the method comprising:

analyzing a first frame among the plurality of frames to determine one or more of a presence and a location of one or more reference objects within the first frame;

determining one or more geometric characteristics of at least one reference object within the first frame;

defining one or more region(s) of interest based at least in part on one or more of the determined geometric characteristics of the at least one reference object within the first frame;

defining one or more subregion(s) of interest within a second frame; and

detecting a presence or an absence of an edge of the document within each defined subregion of interest;

wherein each subregion of interest is defined based at least in part on;

one or more geometric characteristics of at least one of the region(s) of interest within which the subregion of interest was defined; and

one or more of the determined geometric characteristics of the reference object(s) within the first frame.

9. The method as recited in claim 8, wherein the geometric characteristics comprise one or more of:

an alignment angle of a longitudinal axis of a region encompassing the one or more reference object(s);

a height of at least one of the reference object(s);

a baseline corresponding to a plurality of the reference objects; and

an aspect ratio of one or more of the reference objects

an aspect ratio of a region depicting the one or more reference object(s).

10. The method as recited in claim 8, wherein detecting the presence of the edge of the document within each defined region of interest further comprises locating a transition from a background region of the digital image to a non-background region of the digital image within the defined region of interest.

11. The method as recited in claim 10, wherein the detecting further comprises determining a longitudinal axis of the located transition.

12. The method as recited in claim 11, wherein the detecting further comprises determining whether the longitudinal axis of the located transition corresponds to a longitudinal axis of a region encompassing the one or more reference object(s).

13. The method as recited in claim 12, wherein the longitudinal axes are determined to correspond in response to determining one or more of the following conditions are met:

the longitudinal axes are substantially parallel;

the longitudinal axes are substantially perpendicular; and

at least three points along the longitudinal axis of the located transition are each located within a threshold distance of at least three corresponding points along the longitudinal axis of the region encompassing the one or more reference object(s).

14. The method as recited in claim 8, wherein the reference object(s) comprise one or more of magnetic ink character recognition (MICR) characters and machine readable zone (MRZ) characters.

15. The method as recited in claim 8, wherein each subregion of interest is characterized by satisfying one or more of the following criteria:

each subregion of interest is located within a corresponding region of interest defined for the first frame;

each subregion of interest shares at least one boundary with a corresponding region of interest defined for the first frame;

each subregion of interest encompasses an entirety of a corresponding region of interest defined for the first frame;

each subregion of interest encompasses a region bounding a corresponding region of interest defined for the first frame; and

each subregion of interest excludes an area encompassed by a corresponding region of interest defined for the first frame.

16. A computer-implemented method for processing a digital image comprising a digital representation of a document, the method comprising:

determining the document corresponds to one of a predefined set of document types;

analyzing the digital image to determine one or more of a presence and a location of one or more reference objects;

determining one or more geometric characteristics of at least one of the reference objects; and

extrapolating a location of one or more edges of the document within the digital image based at least in part on the determined document type and the one or more geometric characteristics.

17. The method as recited in claim 16, further comprising:

defining one or more region(s) of interest based at least in part on the extrapolated document edge location(s); and

detecting a presence or an absence of a true edge of the document within each defined region of interest.

18. The method as recited in claim 17, wherein the region(s) of interest are further defined based at least in part on one or more of the determined geometric characteristics.

19. The method as recited in claim 16, wherein the reference object(s) comprise one or more of magnetic ink character recognition (MICR) characters and machine readable zone (MRZ) characters.

20. The method as recited in claim 16, wherein the geometric characteristics comprise one or more of:

an alignment angle of a longitudinal axis of a region encompassing the one or more reference object(s);

a height of at least one of the reference object(s);

a baseline corresponding to a plurality of the reference objects; and

an aspect ratio of one or more of the reference objects

an aspect ratio of a region depicting the one or more reference object(s).