Digital point-of-sale analyzer

• US 10,078,826 B2
• Filed: 07/10/2015
• Issued: 09/18/2018
• Est. Priority Date: 08/31/2007
• Status: Active Grant

• Current Assignee: Accenture Global Services Limited (Accenture PLC)
• Original Assignee: Accenture Global Services Limited (Accenture PLC)
• Inventors: Morandi, Simone, Dini, Marco, Sassano, Michele, Campari, Pier Paolo, Fano, Andrew E.
• Primary Examiner(s): Zeender, Florian M
• Assistant Examiner(s): Amsdell, Dana
• Application Number:
  US14/796,709
• Publication Number:
  US 20150317620A1
• Time in Patent Office: 1,166 Days
• CPC Class Codes: G06Q 10/087 Inventory or stock management, e.g. order filling, procurement or balancing against orders
• CPC Class Codes: G06Q 20/203 Inventory monitoring
• CPC Class Codes: G06Q 30/06 Buying, selling or leasing transactions

Abstract

A digital point-of-sale system for determining key performance indicators (KPIs) at a point-of-sale includes a product identification unit and a realogram creation unit. The product identification unit is configured to receive a captured image of a product display and to identify products in the captured image by comparing features determined from the captured image to features determined from products templates. The realogram creation unit is configured to create a realogram from the identified products and product templates. A product price KPI unit is configured to identify a product label proximally located to each identified product, and to recognize the product price on each product label. Each product price is compared to a predetermined range of prices to determine whether the product label proximally located to the identified product is a correct product label for the identified product.

Claims

1. A digital point-of-sale system comprising:

a hardware processor; and

a storage device storing machine-readable instructions executed by the hardware processor to;

receive at least one captured image of a product display, the at least one captured image including a plurality of products having one or more defined physical product features;

analyze the at least one captured image by;

identifying one or more shelves of the product display by analyzing a vertical and horizontal gradient of the at least one captured image, and

identifying, using one or more of the defined physical product features for each of the plurality of products, a location of each product of the plurality of products on the one or more shelves in the at least one captured image

identifying, for each product at the identified location, a proximally located product label from the at least one captured image;

electronically recognizing a product price on the proximally located product label; and

determining whether the proximally located product label is a correct label for the product at the identified location by comparing the electronically recognized product price with at least one predetermined price;

generate a realogram from the analysis of the at least one captured image, wherein the realogram includes an electronically-generated diagram of the product display and locations of each of the plurality of products on the one or more shelves of the product display;

compare the realogram to stored planograms to determine whether the realogram matches a stored planogram; and

in response to the realogram matching a stored planogram, compare the realogram and the matching planogram to determine at least one of deviations of desired product locations, addition of unexpected products, and competitor product interference.

2. The digital point-of-sale system of claim 1, wherein to analyze the at least one captured image, the hardware processor is to determine a local luminance mean and a standard deviation of a local luminance for the at least one captured image, and identify products that are out-of-stock from the local luminance mean and the standard deviation of the local luminance.

3. The digital point-of-sale system of claim 1, wherein the hardware processor is to generate an output based upon the comparison of the realogram to the matching planogram, and wherein the output comprises a message that includes a remedial action with respect to the product display.

4. The digital point-of-sale system of claim 1, wherein to determine whether the realogram matches a stored planogram the hardware processor is to:

determine from the generated realogram quantitative key performance indicators (KPIs) pertaining to a layout of the product display; and

determine, based on the determined quantitative KPIs, whether the generated realogram matches the stored planogram, wherein the stored planogram comprises a list of products identified by product position and orientation.

5. The digital point-of-sale system of claim 4, wherein to determine at least one of the deviations of desired product locations, the addition of unexpected products, and competitor product interference, the hardware processor is to determine qualitative KPIs describing a quality of the product display based on the matching planogram, wherein the qualitative KPIs relate to shelf compliance, a planogram compliance, quality of display, and competitor analysis.

6. The digital point-of-sale system of claim 5, wherein in response to a determination that the generated realogram does not match a stored planogram, the hardware processor is to compare the realogram to matching guidelines, and determine the qualitative KPIs for the product display based on the comparison of the realogram to the matching guidelines.

7. The digital point-of-sale system of claim 4, wherein to determine the quantitative KPIs, the hardware processor is to determine whether each of the identified products is in a desired orientation, in a desired sequence, or located on a predetermined shelf.

8. The digital point-of-sale system of claim 1, wherein the at least one captured image comprises a plurality of captured images, and to generate the realogram, the hardware processor is to align and merge the plurality of captured images into a single image.

9. The digital point-of-sale system of claim 1, wherein to determine whether the proximally located product label is the correct label for the product at the identified location, the hardware processor is to determine whether the product price on the proximally located product label is within a predetermined range of prices based on historical product pricing data for the product at the identified location.

10. A method for analyzing products at a point-of-sale, the method comprising:

receiving, by a hardware processor, at least one captured image of a product display, the at least one captured image including a plurality of products having one or more defined physical product features;

analyzing, by the hardware processor, the at least one captured image by;

identifying one or more shelves of the product display by analyzing a vertical and horizontal gradient of the at least one captured image; and

identifying, using one or more of the defined physical product features for each of the plurality of products, a location of each product of the plurality of products on the one or more shelves in the at least one captured image;

identifying, for each product of the plurality of products, a proximally located product label from the at least one captured image;

electronically recognizing a product price on the proximally located product label; and

determining whether the proximally located product label is a correct label for each product by comparing the electronically recognized product price with at least one predetermined price;

generating a realograrm from the analysis of the at least one captured image, wherein the realogram includes an electronically-generated diagram of the product display and locations of each of the plurality of products on the one or more shelves of the product display;

comparing the realogram to stored planograms to determine whether the realogram matches a stored planogram; and

in response to the realogram matching a stored planogram, comparing the realograrm and the matching planogram to determine at least one of deviations of desired product locations, addition of unexpected products, and competitor product interference.

11. The method of claim 10, wherein analyzing the at least one captured image includes determining a local luminance mean and a standard deviation of a local luminance for the at least one captured image, and identifying products that are out-of-stock from the local luminance mean and the standard deviation of the local luminance.

12. The method of claim 10, comprising:

generating an output based upon the comparing of the realogram to the matching planogram, and wherein generating the output includes generating a message that includes a remedial action with respect to the product display.

13. The method of claim 10, wherein comparing the realogram to the stored planograms to determine whether the realogram matches a stored planogram the hardware processor comprises:

determining, from the generated realogram, quantitative key performance indicators (KPIs) pertaining to a layout of the product display, and

determining, based on the determined quantitative KPIs, whether the generated realogram matches the stored planogram, wherein the stored planogram comprises a list of products identified by product position and orientation.

14. The method of claim 13, wherein the determining of at least one of the deviations of desired product locations, the addition of unexpected products, and competitor product interference, comprises:

determining qualitative KPIs describing a quality of the product display based on the matching planogram, wherein the qualitative KPIs relate to shelf compliance, a planogram compliance, quality of display, and competitor analysis.

15. The method of claim 14, comprising:

in response to determining that the generated realogram does not match a stored planogram, comparing the realogram to matching guidelines and determining the qualitative KPIs pertaining to the layout of the product display based on the comparison of the generated realogram to the matching guidelines.

16. The method of claim 13, wherein determining the quantitative KPIs includes determining whether each of the identified products is in a desired orientation, in a desired sequence, or located on a predetermined shelf.

17. The method of claim 10, wherein the at least one captured image comprises a plurality of captured images, and generating the realogram includes aligning and merging the plurality of captured images into a single image.

18. A non-transitory computer-readable medium comprising machine readable instructions that when executed by a processor, cause the processor to:

receive at least one captured image of a product display, the at least one captured image including a plurality of products having one or more defined physical product features;

analyze the at least one captured image by;

identifying one or more shelves of the product display by analyzing a vertical and horizontal gradient of the at least one captured image, and

identifying, using one or more of the defined physical product features for each of the plurality of products, a location of each product of the plurality of products on the one or more shelves in the at least one captured image;

identifying for each of the plurality of products, a proximally located product label from the at least one captured image;

electronically recognizing a product price on the proximally located product label; and

determining whether the proximally located product label is a correct label for the product at the identified location by comparing the electronically recognized product price with at least one predetermined price;

generate a realogram from the analysis of the at least one captured image, wherein the realogram includes an electronically-generated diagram of the product display and locations of each of the plurality of products on the one or more shelves of the product display;

compare the realogram to stored planograms to determine whether the realogram matches a stored planogram; and

in response to the realogram matching a stored planogram, compare the realogram and the matching planogram to determine at least one of deviations of desired product locations, addition of unexpected products, and competitor product interference.

19. The non-transitory computer-readable medium of claim 18, wherein to analyze the at least one image, the instructions are to cause the processor to determine a local luminance mean and a standard deviation of a local luminance for the at least one image, and identify products that are out-of-stock from the local luminance mean and the standard deviation of the local luminance.