Systems and methods for processing a digital captured image

• US 7,301,564 B2
• Filed: 07/17/2002
• Issued: 11/27/2007
• Est. Priority Date: 07/17/2002
• Status: Expired due to Fees

• Current Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Original Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Inventors: Fan, Jian
• Primary Examiner(s): Ye; Lin
• Assistant Examiner(s): Hannett; James M.
• Application Number:
  US10/197,072
• Publication Number:
  US 20040012679A1
• Time in Patent Office: 1,959 Days
• Field of Search: 348/222.1, 348/239, 348/241, 382/289, 382/293, 382/295, 382/301
• US Class Current: 348/222.1
• CPC Class Codes: G06T 2207/10008 from scanner, fax or copier
• CPC Class Codes: G06T 2207/20061 Hough transform
• CPC Class Codes: G06T 2207/20192 Edge enhancement; Edge preservation
• CPC Class Codes: G06T 2207/30176 Document
• CPC Class Codes: G06T 5/20 using local operators
• CPC Class Codes: G06T 5/70 Denoising; Smoothing
• CPC Class Codes: G06T 5/73 Deblurring; Sharpening
• CPC Class Codes: G06T 7/13 Edge detection
• CPC Class Codes: H04N 1/387 Composing, repositioning or otherwise geometrically modifying originals

Abstract

In one embodiment, the present invention is directed to a method for processing a digitally captured image that comprises an imaged document. The method comprises: detecting graphical information related to spatial discontinuities of the digitally captured image; detecting lines from the detected graphical information; computing effective area parameters for quadrilaterals associated with ones of the detected lines, wherein each effective area parameter for a respective quadrilateral equals an area of the respective quadrilateral modified by at least a corner matching score that is indicative of a number of connected edge pixels in corners of the respective quadrilateral; and selecting a quadrilateral of the quadrilaterals that possesses a largest effective area parameter.

Claims

1. A method for processing a digitally captured image that comprises an imaged document, said method comprising:

detecting graphical information related to spatial discontinuities of said digitally captured image;

detecting lines from said detected graphical information;

computing effective area parameters for quadrilaterals associated with ones of said detected lines, wherein each effective area parameter for a respective quadrilateral equals an area of the respective quadrilateral modified by at least a corner matching score that is calculated from a sum of a number of connected edge pixels and a longest consecutive segment of detected lines in corners of said respective quadrilateral minus the square root of a sum of distances corresponding to the longest consecutive segment divided by a size of the corners; and

selecting a quadrilateral of said quadrilaterals that possesses a largest effective area parameter.

2. The method of claim 1 wherein said computing comprises further modifying the area of each respective quadrilateral by an edge matching score that is indicative of a number of connected edge pixels of sides of said respective quadrilateral.

3. The method of claim 1 further comprising:

performing perspective enhancement of said imaged document utilizing at least said selected quadrilateral.

4. The method of claim 3 wherein said performing perspective enhancement comprises:

mapping a first area in said quadrilateral to a second area in a rectangle of a corrected image data structure, wherein said rectangle comprises graphical information that is substantially limited to said imaged document.

5. The method of claim 1 further comprising:

enhancing graphical information related to spatial continuities of said digitally captured image.

6. The method of claim 1 further comprising:

down-sampling said digitally captured image before performing said detecting.

7. The method of claim 6 further comprising:

mapping down-sampled corners of said selected quadrilateral to pixels associated with said digitally captured image.

8. The method of claim 7 wherein said detecting lines comprises:

determining corner locations of said imaged document utilizing pixels mapped from said down-sampled corners.

21. The method of claim 1, further comprising:

calculating the corner matching score, wherein the corner matching score for each corner equals the sum of the number of connected edge pixels and the longest consecutive segment of the detected lines within the corner minus the square root of a sum of distances corresponding to the longest consecutive segment divided by twice the size of the corners.

9. A system for processing a digitally captured image that comprises an imaged document, said system comprising:

means for performing edge detection on said digitally captured image to produce an edge-enhanced image data structure;

means for detecting lines from said edge-enhanced image data structure;

means for computing effective area parameters for quadrilaterals associated with ones of said detected lines, wherein each effective area parameter for a respective quadrilateral equals on area of the respective quadrilateral modified by at least a corner matching score that is calculated from a sum of a number of connected edge pixels and a longest consecutive segment of detected lines in corners of said respective quadrilateral minus the square root of a sum of distances corresponding to the longest consecutive segment divided by a size of the corners; and

means for selecting a quadrilateral associated with a largest effective area parameter.

10. The system of claim 9 wherein said means for computing further modifies the area of each respective quadrilateral by an edge matching score that is indicative of a number of connected edge pixels of sides of said respective qundrilateral.

11. The method claim 9 comprising:

means for performing perspective enhancement of said imaged document utilizing, at least, said selected quadrilateral.

12. The method of claim 11 wherein said means for performing perspective enhancement is operable to map a first area in said selected quadrilateral to an area of an enhanced image data structure.

13. The system of claim 9 wherein said means for performing edge detection is operable to low-pass filter said digitally captured image.

14. The system of claim 9 further comprising:

means for down-sampling said digitally captured image thut is operable before said means for performing edge detection is operable.

15. The system of claim 14 further comprising:

mapping down-sampled corners of said selected quadrilateral to pixels in said digitally captured image.

16. The system of claim 9 further comprising:

means for transforming said digitally captured image to luminance image before said means for performing edge detection is operable.

17. A computer-readable medium encoded with a data structure comprising executable instructions for processing a digitally captured image, said data structure comprising:

code for detecting graphical information associated with spatial discontimiities of said digitally captured image;

code for detecting lines from said detected graphical information;

code for computing effective area parameters for quadrilaterals associated with ones of said detected lines, wherein each effective area parameter for a respective quadrilateral equals an area of the respective quadrilateral modified by at least a corner matching score that is calculated from a sum of a number of connected edge pixels and a longest consecutive segment of detected lines in corners of said respective quadrilateral minus the square root of a sum of distances corresponding to the longest consecutive segment divided by a size of the corners; and

code for selecting a quadrilateral of said possible quadrilaterals that possesses a largest effective area parameter.

18. The computer-readable medium of claim 17 wherein said code for computing further modifies the area of each respective quadrilateral by an edge matching score that is indicative of a number of connected edge pixels of sides of said respective quadrilateral.

19. The computer-readable medium of claim 17 further comprising;

code for performing perspective enhancement of said image document utilizing at least said selected quadrilateral.

20. The computer-readable medium of claim 17 further comprising:

code for removing respective edge pixels wherein each of said respective edge pixel is associated with a neighborhood that possesses at least one item from the list of;

(1) a number of pixels that is less than a first threshold and (2) a number of pixels that are greater than a second threshold.