Method and apparatus for rapid and precision detection of omnidirectional postnet barcode location

• US 6,708,884 B1
• Filed: 06/25/2002
• Issued: 03/23/2004
• Est. Priority Date: 06/25/2002
• Status: Expired due to Fees

• Current Assignee: United States Of America As Represented By The Secretary Of The Army (Government of the United States of America)
• Original Assignee: The United States Of America As Represented By The Secretary Of The Army
• Inventors: Shi, Yun-Qing, Su, Wei, Chang, Cheng-jyh, Lin, Shu
• Primary Examiner(s): Frech, Karl D.
• Assistant Examiner(s): FUREMAN, JARED
• Application Number:
  US10/319,670
• Time in Patent Office: 637 Days
• Field of Search: 235/462.09, 235/462.08, 235/462.1, 235/462.11, 235/462.02, 235/462.04, 235/462.25, 235/462.41, 235/462.01, 235/454, 235/470, 382/101, 382/102, 382/181, 382/190, 382/209, 382/289
• US Class Current: 235/462.09
• CPC Class Codes: G06K 7/14 using light without selection of wavelength, e.g. sensing reflected white light G06K7/10831 - G06K7/1097 take precedence
• CPC Class Codes: G06K 7/1443 locating of the code in an image

Abstract

A method for rapid and precise detecting of an omnidirectional postnet bar code on an object by digital signal processing is provided. In order to rapidly and accurately detect omnidirectional postnet bar codes, a two-dimensional digital image containing the postnet bar code is acquired, filtered and dilated to form a block dominated by a plurality of black-colored pixels. Then a down sampled image is provided and match filtered with a two-dimensional matched filtering output to indicate a best-matched filter, thus determining a postnet bar code location and orientation by associating matched filters with an orientation angle of the omnidirectional postnet bar code. The method further comprises the steps of identifying a gravity center for each short bar code and connecting the gravity centers to form a straight line by using a Hough transform and comparing the straight line with the postnet bar code location to generate a verification result and then detecting a position and an orientation of said postnet bar code location by matching said postnet bar code location with said verification result.

Claims

1. A method for detecting an omnidirectional postnet bar code on an object by digital signal processing, comprising the steps of:

acquiring a two-dimensional digital image of said omnidirectional postnet bar code, said two-dimensional digital image being represented by a plurality of color-coded pixels;

filtering said two-dimensional digital image with a means for filtering to remove a plurality of non-black coded pixels from said plurality of color-coded pixels and provide a filtering output;

saving said filtering output in a memory device;

dilating said two-dimensional digital image to form a block dominated by a plurality of black-colored pixels;

providing a down sampled image G_m;

match-filtering said down-sampled image G_m with a set of two-dimensional matched filters to indicate a best-matched filter;

determining a an omnidirectional postnet bar code location;

associating a plurality of matched filters with an orientation angle of said omnidirectional postnet bar code;

generating a plurality of coefficients from said match-filtering step during a feature template step;

identifying a gravity center for each of a plurality of short bar codes and connecting said gravity centers to form a straight line;

comparing said straight line with said omnidirectional postnet bar code location to generate a verification result; and

detecting a position and an orientation of said omnidirectional postnet bar code location by matching said omnidirectional postnet bar code location with said verification result.

2. The method for detecting an omnidirectional postnet bar code on an object by digital signal processing, as recited in claim 1, further comprising said filtering output including an M×

N binary data array denoted by a G_f factor from said filtering step.

3. The method for detecting an omnidirectional postnet bar code on an object by digital signal processing, as recited in claim 2, further comprising the step of assigning a binary number 0 to said plurality of non-black coded pixels.

4. The method for detecting an omnidirectional postnet bar code on an object by digital signal processing, as recited in claim 3, further comprising the step of assigning a binary number 1 to said plurality of black-coded pixels.

5. The method for detecting an omnidirectional postnet bar code on an object by digital signal processing, as recited in claim 4, wherein said block provides a rectangular block pattern for an image recognition pattern.

6. The method for detecting an omnidirectional postnet bar code on an object by digital signal processing, as recited in claim 5, further comprising the step of down sampling said two-dimensional digital image with a multi-grid data processing means.

7. The method for detecting an omnidirectional postnet bar code on an object by digital signal processing, as recited in claim 6, further comprising the step of match filtering said down-sampled image G_m with a plurality of pipelines.

8. The method for detecting an omnidirectional postnet bar code on an object by digital signal processing, as recited in claim 7, further comprising the step of said best-matched filter determining said omnidirectional postnet bar code location.

9. The method for detecting an omnidirectional postnet bar code on an object by digital signal processing, as recited in claim 8, further comprising the step of a non-linear rotation during said feature template step.

10. The method for detecting an omnidirectional postnet bar code on an object by digital signal processing, as recited in claim 9, further comprising the step of a linear rotation during said feature template step.

11. The method for detecting an omnidirectional postnet bar code on an object by digital signal processing, as recited in claim 9, further comprising the step of forming said straight line by a Hugh transform.

12. The method for detecting an omnidirectional postnet bar code on an object by digital signal processing, as recited in claim 11, further comprising the step of acquiring said two-dimensional digital image by scanning said object.

13. The method for detecting an omnidirectional postnet bar code on an object by digital signal processing, as recited in claim 12, further comprising the step of eliminating noise in said two-dimensional digital image during the dilation step.

14. The method for detecting an omnidirectional postnet bar code on an object by digital signal processing, as recited in claim 13, further comprising the step of eliminating a plurality of voids in said two-dimensional digital image during the dilation step.

15. The method for detecting an omnidirectional postnet bar code on an object by digital signal processing, as recited in claim 14, further comprising said best-matched filter being determined by the formulas:

16. The method for detecting an omnidirectional postnet bar code on an object by digital signal processing, as recited in claim 15, further comprising the step of calculating a plurality of moments and a plurality of inertials from the plurality of short bar codes.

17. The method for detecting an omnidirectional postnet bar code on an object by digital signal processing, as recited in claim 16, wherein packages are rapidly sorted into zip codes.

18. The method for detecting an omnidirectional postnet bar code on an object by digital signal processing, as recited in claim 17, further comprising automating said steps with computer equipment.

19. The method for detecting an omnidirectional postnet bar code on an object by digital signal processing, as recited in claim 11, further comprising the step of acquiring said two-dirnensional digital image by photographing said object.

20. A computer-readable medium whose contents cause a computer system to detect an omnidirectional postnet bar code on an object by digital signal processing, comprising:

a means for image acquisition generates a two-dimensional digital image of said omnidirectional postnet bar code;

said digital image further comprising a plurality of color-coded pixels;

said omnidirectional postnet bar code having a plurality of short bars;

a means for filtering removes a plurality of non-black coded pixels from said plurality of color-coded pixels to provide a filtering output, said filtering output being saved to a memory device of said computer system;

a means for dilating said two-dimensional digital image forms a block dominated by a plurality of black-colored pixels;

a means for down-sampling provides a down sampled image G_m;

a means for matched filtering processes said down-sampled image G_m to provide a best-matched filter;

said best-matched filter determines said omnidirectional postnet bar code location;

said matched filtering means associates a plurality of matched filters with an orientation angle of said omnidirectional postnet bar code;

a feature template means generates a plurality of coefficients;

each of said plurality of short bar codes having a gravity center;

a means for Hugh transform connects said gravity centers to form a straight line, said straight line being compared with said omnidirectional postnet bar code location to generate said verification result; and

said verification result is matched with said omnidirectional postnet bar code location by a means for orientation to detect a position and said orientation of said omnidirectional postnet bar code location.

21. The computer-readable medium whose contents cause a computer system to detect an omnidirectional postnet bar code on an object by digital signal processing, as recited in claim 20, further comprising said filtering output including an M×

N binary data array denoted by a G_f factor from said filtering means.

22. The computer-readable medium whose contents cause a computer system to detect an omnidirectional postnet bar code on an object by digital signal processing, as recited in claim 21, further comprising said block provides a rectangular block pattern for an image recognition pattern.

23. The computer-readable medium whose contents cause a computer system to detect an omnidirectional postnet bar code on an object by digital signal processing, as recited in claim 22, further comprising down sampling said two-dimensional digital image with a multi-grid data processing means.

24. The computer-readable medium whose contents cause a computer system to detect an omnidirectional postnet bar code on an object by digital signal processing, as recited in claim 23, further comprising said match filtering means match filters said down-sampled image G_m with a set of two-dimensional matched filters to indicate said best-matched filter.

25. The computer-readable medium whose contents cause a computer system to detect an omnidirectional postnet bar code on an object by digital signal processing, as recited in claim 24, further comprising a non-linear rotation after said plurality of coefficients are generated.

26. The computer-readable medium whose contents cause a computer system to detect an omnidirectional postnet bar code on an object by digital signal processing, as recited in claim 25, further comprising transferring said contents of the computer-readable medium over a computer network.

27. The computer-readable medium whose contents cause a computer system to detect an omnidirectional postnet bar code on an object by digital signal processing, as recited in claim 26, further comprising said best-matched filter being determined by the formulas:

28. The computer-readable medium whose contents cause a computer system to detect an omnidirectional postnet bar code on an object by digital signal processing, as recited in claim 27, wherein packages are rapidly sorted into zip codes.

29. The computer-readable medium whose contents cause a computer system to detect an omnidirectional postnet bar code on an object by digital signal processing, as recited in claim 24, further comprising a linear rotation after said plurality of coefficients are generated.

30. A storage medium encoded with machine-readable computer program code whose contents cause a computer system to detect a position and orientation of an omnidirectional postnet bar code on an object by digital signal processing to match a omnidirectional postnet bar code location with a verification result, comprising:

a means for image acquisition generates a two-dimensional digital image of said omnidirectional postnet bar code;

said two-dimensional digital image further comprising a plurality of color-coded pixels;

said omnidirectional postnet bar code having a plurality of short bar bars;

a means for filtering removes a plurality of non-black coded pixels from said plurality of color-coded pixels to provide a filtering output, said filtering output being saved to a memory device of said computer system;

a means for dilating said two-dimensional digital image forms a block dominated by a plurality of black-colored pixels;

a means for multi-grid image processing provides a down sampled image G_m;

a means for matched filtering match filters said down-sampled image G_m with a set of two-dimensional matched filters provides a best-matched filter;

said best-matched filter determines said omnidirectional postnet bar code location;

said matched filtering means associates a plurality of matched filters with an orientation angle of said omnidirectional postnet bar code;

a feature template means generates a plurality of coefficients;

each of said plurality of short bar codes having a gravity center;

a means for Hugh transform connects said gravity centers to form a straight line, said straight line being compared with said omnidirectional postnet bar code location to generate said verification result; and

said verification result is matched with said omnidirectional postnet bar code location to detect a position and said orientation of said omnidirectional postnet bar code location.

31. The storage medium encoded with machine-readable computer program code whose contents cause a computer system to detect a position and orientation of an omnidirectional postnet bar code, as recited in claim 30, further comprising said filtering output including a M×

N binary data array denoted by a G_f factor from said filtering means.

32. The storage medium encoded with machine-readable computer program code whose contents cause a computer system to detect a position and orientation of an omnidirectional postnet bar code, as recited in claim 31, further comprising said block provides a rectangular block pattern for an image recognition pattern.

33. The storage medium encoded with machine-readable computer program code whose contents cause a computer system to detect a position and orientation of an omnidirectional postnet bar code, as recited in claim 32, further comprising down sampling said two-dimensional digital image with a multi-grid data processing means.

34. The storage medium encoded with machine-readable computer program code whose contents cause a computer system to detect a position and orientation of an omnidirectional postnet bar code, as recited in claim 33, further comprising said match filtering means filters said down sampled image G_m with a set of two-dimensional matched filters indicating said best-matched filter.

35. The storage medium encoded with machine-readable computer program code whose contents cause a computer system to detect a position and orientation of an omnidirectional postnet bar code, as recited in claim 34, further comprising a non-linear rotation after said plurality of coefficients are generated.

36. The storage medium encoded with machine-readable computer program code whose contents cause a computer system to detect a position and orientation of an omnidirectional postnet bar code, as recited in claim 35, further comprising transferring said machine-readable computer program code over a computer network.

37. The storage medium encoded with machine-readable computer program code whose contents cause a computer system to detect a position and orientation of an omnidirectional postnet bar code, as recited in claim 36, further comprising said best-matched filter being determined by the formulas:

38. The storage medium encoded with machine-readable computer program code whose contents cause a computer system to detect a position and orientation of an omnidirectional postnet bar code, as recited in claim 37, wherein packages are rapidly sorted into zip codes.

39. The storage medium encoded with machine-readable computer program code whose contents cause a computer system to detect a position and orientation of an omnidirectional postnet bar code, as recited in claim 34, further comprising a linear rotation after said plurality of coefficients are generated.