IMAGE PROCESSING METHODS AND SYSTEMS FOR BARCODE AND/OR PRODUCT LABEL RECOGNITION

• US 20150363625A1
• Filed: 06/13/2014
• Published: 12/17/2015
• Est. Priority Date: 06/13/2014
• Status: Active Grant

• Current Assignee: Conduent Business Services, LLC (Conduent, Inc.)
• Original Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Inventors: Wu, Wencheng, Venable, Dennis L., Moore, Steven R., Wade, Thomas F., Paul, Peter, Ct, Adrien P.
• Granted Patent: US 9,659,204 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• 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 7/146 the method including quality enhancement steps
• CPC Class Codes: G06V 10/225 based on a marking or identifier characterising the area
• CPC Class Codes: G06V 20/10 Terrestrial scenes scenes under surveillance with static cameras G06V20/52; scenes perceived from the exterior of a vehicle G06V20/56; scenes perceived from the interior of a vehicle G06V20/59
• CPC Class Codes: G06V 20/52 Surveillance or monitoring of activities, e.g. for recognising suspicious objects recognising microscopic objects G06V20/69

Abstract

This disclosure provides an image processing method and system for recognizing barcodes and/or product labels. According to an exemplary embodiment, the method uses a multifaceted detection process that includes both image enhancement of a candidate barcode region and other product label information associated with a candidate barcode region to identify a product label, where the candidate barcode region includes a nonreadable barcode. According to one exemplary application, a store profile is generated based on the identifications of the product labels which are associated with a location of a product within a store.

Claims

1. A method of performing decoding of a barcode associated with a product label, the product label including one or more barcodes and other product label information, the method comprising:

an image capturing device capturing an image of the product label and storing the captured image in a memory operatively associated with the image capturing device;

a processor operatively associated with the memory detecting and localizing one or more barcode candidate regions within the captured image of the product label, the barcode candidate regions including a substantially fewer number of pixels relative to a total number of pixels included in the captured pixel image of the product label;

cropping the detected and localized one or more barcode candidate regions to generate one or more sub-images including images of the one or more barcode candidate regions;

processing each sub-image using two or more independent image quality improvement processes to generate modified versions of the barcode candidate regions; and

processing the modified versions of the barcode candidate regions to decode readable barcodes included in the modified versions of the candidate barcode regions.

2. The method of performing decoding of a barcode according to claim 1, wherein the two or more independent image quality improvement processes include one or more of contrast enhancement, tone-curve reshaping, gamma-correction, sharpening, de-blurring and morphological filtering.

3. The method of performing decoding of a barcode according to claim 1, wherein the two or more independent image quality improvement processes are performed in serial or parallel.

4. The method of performing decoding of a barcode associated with a product label according to claim 1, wherein the barcode is one of a 1D (Dimensional) and 2D barcode.

5. The method of performing decoding of a barcode associated with a product label according to claim 1, further comprising:

associating the product label including the decoded barcode with a location in a retail environment.

6. The method of performing decoding of a barcode associated with a product label according to claim 5, wherein the step of detecting and localizing one or more barcode-candidate-regions is biased to generate relatively more false-positives than false-negatives.

7. The method of performing decoding of a barcode associated with a product label according to claim 1, further comprising:

fusing the decoded readable barcodes and identifying a product label based on the fusion.

8. The method of performing decoding of a barcode associated with a product label according to claim 1, further comprising:

further processing the captured image of the product label to extract other non-barcode information to identify the product label if an associated candidate-barcode-region includes an unreadable barcode.

22. The method of performing product label identification according to claim 1, wherein the product label identification includes a SKU (stock keeping unit).

23. The method of performing product label identification according to claim 1, wherein the other product label information includes one or more SKU text, price text, product information text, logo, and color patch.

9. An image processing system for performing decoding of a barcode associated with a product label, the product label including one or more barcodes and other product label information, the image processing system comprising:

an image capturing device; and

a processor operatively connected to the image capturing device, the processor configured to;

the image capturing device capturing an image of the product label and storing the captured image in a memory operatively associated with the image capturing device;

the processor operatively associated with the memory detecting and localizing one or more barcode candidate regions within the captured image of the product label, the barcode candidate regions including a substantially fewer number of pixels relative to a total number of pixels included in the captured pixel image of the product label;

cropping the detected and localized one or more barcode candidate regions to generate one or more sub-images including images of the one or more barcode candidate regions;

processing each sub-image using two or more independent image quality improvement processes to generate modified versions of the barcode candidate regions; and

processing the modified versions of the barcode candidate regions to decode readable barcodes included in the modified versions of the candidate barcode regions.

10. The image processing system according to claim 9, wherein the two or more independent image quality improvement processes include one or more of contrast enhancement, tone-curve reshaping, gamma-correction, sharpening, de-blurring and morphological filtering.

11. The image processing system according to claim 9, wherein the two or more independent image quality improvement processes are performed in serial or parallel.

12. The image processing system according to claim 9, wherein the barcode is one of a 1D (Dimensional) and 2D barcode.

13. The image processing system according to claim 9, further comprising:

associating the product label including the decoded barcode with a location in a retail environment.

14. The image processing system according to claim 13, wherein the step of detecting and localizing one or more barcode-candidate-regions is biased to generate relatively more false-positives than false-negatives.

15. The image processing system according to claim 9, further comprising:

fusing the decoded readable barcodes and identifying a product label based on the fusion.

16. The image processing system according to claim 9, further comprising:

further processing the captured image of the product label to extract other non-barcode information to identify the product label if an associated candidate-barcode-region includes an unreadable barcode.

17. A method of performing product label identification, the product label including one or more barcodes and other product label information, the method comprising:

an image capturing device capturing an image of the product label and storing the captured image in a memory operatively associated with the image capturing device;

a processor operatively associated with the memory detecting and localizing one or more barcode candidate regions within the captured image of the product label;

cropping the detected and localized one or more barcode candidate regions to generate one or more sub-images including the one or more barcode candidate regions;

processing each sub-image to decode readable barcodes included in the barcode candidate regions and identify barcode candidate regions including an unreadable barcode;

processing all or a portion of the captured image of the product label associated with the unreadable barcode to determine all or part of the other product label information association with the unreadable barcode;

comparing the determined other product label information to a plurality of product label templates to determine a layout associated with the product label including an unreadable barcode candidate region;

processing the captured image of the product label to extract all or part of the other product label information based on the determined layout associated with the product label; and

identifying the captured image of the product label as including one of a plurality of unique predefined product labels.

18. The method of performing product label identification according to claim 17, wherein an estimated orientation of a barcode candidate region associated with a nonreadable barcode is used to apply a correction to all or part of the captured image of the product label.

19. The method of performing product label identification according to claim 18, wherein the correction is associated with one or more of skew and rotation.

20. The method of performing product label identification according to claim 17, wherein the step of processing the captured image of the product label selects and uses one of a plurality of extraction algorithms based on the determined layout associated with the product label.

21. The method of performing product label identification according to claim 17, wherein the step of identifying the captured image of the product label includes associating a relative confidence value with the identification of the product label.