Checkout terminal

Patent metadata

Field	Value
Publication number	US-12444278-B2
Application number	US-202218696303-A
Country	US
Kind code	B2
Filing date	Sep 22, 2022
Priority date	Sep 27, 2021
Publication date	Oct 14, 2025
Grant date	Oct 14, 2025

Abstract

A checkout terminal is provided. The checkout terminal comprises a camera array, a display device, a weight scale, and a control circuit. The control circuit is configured to identify, based at least on images captured by the camera array, product identifiers associated with a plurality of different items placed in a placement area and on the weight scale, determine a combined weight of the plurality of different items based on product weight information stored in a product database, retrieve a weight measurement measured by the weight scale, detect for unaccounted items based on comparing the combined weight the plurality of different items and the weight measurement, and in the event that an unaccounted item is detected, display, via the display device, instructions to move one or more items to identify the unaccounted item.

First claim

What is claimed is:

1. A checkout terminal system comprising:
a camera array;
a display device;
a weight scale; and
a control circuit coupled to the camera array, the display device, and the weight scale, the control circuit being configured to:
identify, based at least on images captured by the camera array, product identifiers associated with a plurality of different items placed in a placement area of the weight scale;
identify an item in the plurality of different items as a variable weight item;
cause the display device to display instructions to lift the variable weight item;
determine a weight of the variable weight item based on a change in a total weight on the weight scale when the variable weight item is lifted;
determine a cost of the variable weight item based on the weight;
determine a combined weight of the plurality of different items based on item weight information stored in a product database;
retrieve a weight measurement of items in the placement area from the weight scale;
detect for unaccounted items in the placement area based on comparing the combined weight of the plurality of different items and the weight measurement of items in the placement area;
in the event that an unaccounted item is detected in the placement area, display, via the display device and while the items are in the placement area, instructions to move one or more items in the placement area; and
identify, based on at least additional images captured by the camera array, the unaccounted item.

2. The system of claim 1 , wherein the camera array comprises a plurality of 2D cameras and a plurality of depth cameras providing different fields of view around the placement area.

3. The system of claim 1 , wherein the camera array comprises a plurality of cameras embedded in the weight scale.

4. The system of claim 1 , further comprising:
a first camera support housing a first depth camera, a first 2D cameras, and a second 2D camera of the camera array.

5. The system of claim 4 , further comprising:
a second camera support housing a second depth camera, a third 2D cameras, and a fourth 2D camera of the camera array.

6. The system of claim 1 , further comprising:
a Radio Frequency Identification (RFID) reader embedded in the weight scale,
wherein a subset of the plurality of different items is identified via the RFID reader.

7. The system of claim 1 , wherein a subset of the plurality of different items are identified based on detecting optically readable codes on items in images captured by the camera array.

8. The system of claim 1 , wherein a subset of the plurality of different items are identified via a machine learning algorithm based on a computer vision model trained on images captured at a plurality of checkout terminals.

9. The system of claim 8 , wherein the images captured by the camera array and identifiers associated with the plurality of different items are used to train the computer vision model for future identifications.

10. The system of claim 1 , further comprising:
an overhead unit positioned over the weight scale, the overhead unit comprises one or more cameras of the camera array.

11. The system of claim 1 , further comprising:
a light source over the weight scale, wherein the control circuit is configured to control the light source to adjust a lighting condition of the plurality of different items placed on the weight scale.

12. The system of claim 1 , further comprising:
a second display device;
wherein the control circuit is configured to cause information associated with identified items to be displayed on the second display device.

13. The system of claim 1 , further comprising:
a card reader configured to accept payment for a purchase of the plurality of different items.

14. A retail checkout method comprising:
identifying, at a control circuit and based at least on images captured by a camera array of a checkout terminal, product identifiers associated with a plurality of different items placed in a placement area of a weight scale of the checkout terminal;
identifying an item in the plurality of different items as a variable weight item;
causing a display device to display instructions to lift the variable weight item;
determining, by the control circuit, a weight of the variable weight item based on a change in a total weight on the weight scale when the variable weight item is lifted;
determining, by the control circuit, a cost of the variable weight item based on the weight;
determining, at the control circuit, a combined weight of the plurality of different items based on item weight information stored in a product database;
retrieving, by the control circuit, a weight measurement of items in the placement area from the weight scale;
detecting, by the control circuit, for unaccounted items in the placement area based on comparing the combined weight of the plurality of different items and the weight measurement of items in the placement area;
in the event that an unaccounted item is detected in the placement area, displaying, via the display device of the checkout terminal and while the items are in the placement area, instructions to move one or more items in the placement area; and
identifying, based on at least additional images captured by the camera array, the unaccounted item.

15. The method of claim 14 , wherein the camera array comprises a plurality of 2D cameras and a plurality of depth cameras providing different fields of view around the placement area.

16. The method of claim 14 , wherein the camera array comprises a plurality of cameras embedded in the weight scale.

17. The method of claim 14 , further comprising:
wherein a subset of the plurality of different items is identified via a Radio Frequency Identification (RFID) reader embedded in the weight scale.

18. The method of claim 14 , wherein a subset of the plurality of different items are identified based on detecting optically readable codes on items in images captured by the camera array.

19. The method of claim 14 , wherein a subset of the plurality of different items are identified via a machine learning algorithm based on a computer vision model trained on images captured at a plurality of checkout terminals.

20. The method of claim 19 , wherein the images captured by the camera array and identifiers associated with the plurality of different items are used to train the computer vision model for future identifications.

Assignees

• Walmart Apollo Llc

Inventors

• Whitelaw Darrell
• Grignon Andrew J
• Pachuta Kristine L
• Cervantes Michael C
• Lew Jae Y
• Hadinger Stephen S
• Leek Brittany N
• Yergin William J
• Chee Wei-Meng
• Crecelius Jr John C

Classifications

• G07G1/01

  Details for indicating · CPC title

• G07G1/009

  the reader being an RFID reader · CPC title

• G06T2207/20084

  Artificial neural networks [ANN] · CPC title

• G06T2207/20081

  Training; Learning · CPC title

• G06T2207/10028

  Range image; Depth image; 3D point clouds · CPC title