SYSTEM AND METHOD FOR RECOGNIZING DEFORMED LINEAR BARCODES FROM A STREAM OF VARIED-FOCUS VIDEO FRAMES

• US 20110068173A1
• Filed: 09/17/2010
• Published: 03/24/2011
• Est. Priority Date: 09/24/2009
• Status: Active Grant

• Current Assignee: eBay Inc.
• Original Assignee: eBay Inc.
• Inventors: Reddy, Vikas Muppiddi, Powers, Jeffrey Roger
• Granted Patent: US 8,851,378 B2
• Time in Patent Office: Days
• US Class Current: 235/462.06
• CPC Class Codes: G06K 5/00 Methods or arrangements for verifying the correctness of markings on a record carrier; Column detection devices
• CPC Class Codes: G06K 7/10821 further details of bar or optical code scanning devices
• 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 7/1413 1D bar codes
• CPC Class Codes: G06K 7/1443 locating of the code in an image
• CPC Class Codes: G06K 7/1447 extracting optical codes from image or text carrying said optical code
• CPC Class Codes: G06K 7/1473 error correction
• CPC Class Codes: G06K 7/1478 adapting the threshold for pixels in a CMOS or CCD pixel sensor for black and white recognition

Abstract

In a system and method of recognizing a barcode from a stream of video frames, a processor-implemented camera module receives a stream of video frames, with at least one video frame including a barcode. A processor-implemented barcode blur estimate module estimates an amount of defocus blur in a video frame. The processor-implemented barcode blur estimate module further estimates an identity of the barcode. A processor-implemented barcode localization module identifies a region of the video frame containing the barcode. A processor-implemented barcode geometric modeler module generates a geometric model of the barcode that includes an identified barcode deformity. A processor-implemented barcode decoder module decodes the barcode from the video frame using the estimated amount of defocus blur, the estimated identity of the barcode, and the geometric model of the barcode.

Claims

1. A system, comprising:

a processor-implemented barcode blur estimation module configured to;

estimate an amount of defocus blur in a video frame of a stream of video frames; and

estimate an identity of a barcode contained in the video frame;

a processor-implemented barcode localization module configured to identify a region of the video frame containing the barcode;

a processor-implemented barcode geometric modeler module configured to generate a geometric model of the barcode, the geometric model of the barcode including an identified barcode deformity; and

a processor-implemented barcode decoder module configured to decode the barcode in the video frame using the estimated barcode identity and the geometric model of the barcode.

2. The system of claim 1, further comprising a processor-implemented camera module configured to provide the stream of video frames to the processor-implemented barcode blur estimation module, wherein at least one video frame contains the barcode.

3. The system of claim 1, wherein the processor-implemented barcode blur estimation module is further configured to perform a backward extraction technique to estimate the identity of the barcode.

4. The system of claim 3, wherein performing the backward extraction technique comprises the processor-implemented blur estimation module being further configured to:

estimate elements of the barcode contained in the video frame;

filter barcode candidates using the estimated barcode elements to obtain a subset of feasible barcode candidates; and

compare the estimated barcode elements to the subset of feasible barcode candidates to determine the identity of the barcode.

5. The system of claim 1, wherein the identified barcode deformity includes at least one of a curved barcode, a skewed barcode, a rotated barcode, and a lighting condition indicating an amount of light in the video frame.

6. The system of claim 1, wherein the processor-implemented barcode decoder module is configured to decode the barcode in the identified region of the video frame.

7. The system of claim 1, further comprising a communication module configured to transmit decoded barcode data and receive information related to a product identified by the decoded barcode data.

8. The system of claim 1, wherein estimating the amount of defocus blur comprises the processor-implemented blur estimation module being further configured to:

extract a cross-section from the barcode contained in the video frame;

calculate a difference in pixel intensity among a set of pixels within the barcode cross-section;

aggregate the difference in pixel intensity among the set of pixels; and

estimate a radius of defocus blur using the aggregated difference in pixel intensity.

9. The system of claim 1, wherein the processor-implemented barcode decoder module is configured to decode the barcode in response to the estimated amount of defocus blur being below a predetermined threshold of blur.

10. A computer-implemented method, comprising:

receiving, at a device having a processor, a stream of video frames from a camera;

estimating an amount of defocus blur in a video frame of the stream of video frames;

estimating an identity of the barcode using a backward extraction technique;

identifying a region of the video frame containing a barcode;

generating a geometric model of the barcode, the geometric model including an identified barcode deformity; and

decoding the barcode from the identified region of the video frame using the estimated amount of defocus blur, the estimated barcode identity, and the geometric model of the barcode.

11. The computer-implemented method of claim 10, wherein the estimating the amount of defocus blur, the estimating the identity of the barcode, the identifying, the generating, and the decoding are performed simultaneously for each video frame of the stream of video frames.

12. The computer-implemented method of claim 10, wherein estimating the amount of defocus blur comprises:

extracting a cross-section from the barcode contained in the video frame;

analyzing a difference in pixel intensity among a set of pixels within the barcode cross-section;

aggregating the difference in pixel intensity among the set of pixels; and

calculating a radius of defocus blur using the aggregated difference in pixel intensity.

13. The computer-implemented method of claim 10, wherein the backward extraction technique comprises:

estimating elements of the barcode contained in the video frame;

filtering barcode candidates using the estimated barcode elements to obtain a subset of feasible barcode candidates; and

comparing the estimated barcode elements to the subset of feasible barcode candidates to determine the identity of the barcode.

14. The computer-implemented method of claim 10, wherein the identified barcode deformity includes at least one of a curved barcode, a skewed barcode, a rotated barcode, and a lighting condition indicating an amount of light in the video frame.

15. The computer-implemented method of claim 10, further comprising:

transmitting the decoded barcode data to a server via a network; and

receiving information related to a product identified by the decoded barcode data.

16. The computer-implemented method of claim 10, wherein the barcode is decoded in response to a determination that the amount of defocus blur is below a predetermined threshold of blur, and wherein in response to a determination that the amount of defocus blur is above the predetermined threshold of blur, the barcode is not decoded.

17. A computer-implemented method, comprising:

receiving, at a device having a processor, a stream of video frames from a camera; and

decoding a barcode contained in the stream of video frames.

18. The computer-implemented method of claim 17, further comprising estimating an amount of defocus blur in the stream of video frames, wherein the barcode is decoded when the amount of defocus blur in the stream of video frames is less than a predetermined threshold of blur.

19. The computer-implemented method of claim 17, further comprising:

transmitting the decoded barcode to a server via a network; and

receiving information related to a product identified by the decoded barcode.

20. A non-transitory machine-readable storage medium that stores a set of instructions that, when executed by a machine, causes the machine to perform operations, comprising:

receiving a stream of video frames from a camera; and

decoding a barcode contained in the stream of video frames.

21. The non-transitory machine-readable storage medium of claim 20, further comprising estimating an amount of defocus blur in the stream of video frames, wherein the barcode is decoded when the amount of defocus blur in the stream of video frames is less than a predetermined threshold of blur.

22. The non-transitory machine-readable storage medium of claim 20, further comprising:

transmitting the decoded barcode to a server via a network; and

receiving information related to a product identified by the decoded barcode.

23. A non-transitory machine-readable storage medium that stores a set of instructions that, when executed by a machine, causes the machine to perform operations, comprising:

receiving, at a device having a processor, a stream of video frames from a camera;

estimating an amount of defocus blur in a video frame of the stream of video frames;

estimating an identity of the barcode using a backward extraction technique;

identifying a region of the video frame containing a barcode;

generating a geometric model of the barcode, the geometric model including an identified barcode deformity; and

decoding the barcode from the identified region of the video frame using the estimated amount of defocus blur, the estimated barcode identity, and the geometric model of the barcode.

24. The non-transitory machine-readable storage medium of claim 23, wherein the estimating the amount of defocus blur, the estimating the identity of the barcode, the identifying, the generating, and the decoding are performed simultaneously for each video frame of the stream of video frames.

25. The non-transitory machine-readable storage medium of claim 23, wherein estimating the amount of defocus blur comprises:

extracting a cross-section from the barcode contained in the video frame;

analyzing a difference in pixel intensity among a set of pixels within the barcode cross-section;

aggregating the difference in pixel intensity among the set of pixels; and

estimating a radius of defocus blur using the aggregated difference in pixel intensity.

26. The non-transitory machine-readable storage medium of claim 23, wherein the backward extraction technique comprises:

estimating elements of the barcode contained in the video frame;

filtering barcode candidates using the estimated barcode elements to obtain a subset of feasible barcode candidates; and

comparing the estimated barcode elements to the subset of feasible barcode candidates to determine the identity of the barcode.

27. The non-transitory machine-readable storage medium of claim 23, wherein the identified barcode deformity includes at least one of a curved barcode, a skewed barcode, a rotated barcode, and a lighting condition indicating an amount of light in the video frame.

28. The non-transitory machine-readable storage medium of claim 23, further comprising:

transmitting the decoded barcode data to a server via a network; and

receiving information related to a product identified by the decoded barcode data.

29. The non-transitory machine-readable storage medium of claim 23, wherein the barcode is decoded in response to a determination that the amount of defocus blur is below a predetermined threshold of blur, and wherein in response to a determination that the amount of defocus blur is above the predetermined threshold of blur, the barcode is not decoded.