IP Library Granted Patent US 10,410,025
Granted Patent B2
US 10,410,025 · App. 15/954,083 · Granted Sep 10, 2019

Product checkout and interest detection in retail environment using radio-frequency identification

View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 10,410,025
App. No.
15/954,083
Granted
Sep 10, 2019
Kind
B2
Abstract

Systems and a method are provided for use with a shopping cart in a store. A system includes Radio-Frequency Identification (RFID) tags, including item RFID tags associated with items for sale and reference RFID tags associated with various aisles in the store, for providing RFID data. The system further includes a RFID reader, operatively coupled to the shopping cart, and configured to the read RFID data from the item RFID tags and the reference RFID tags, extract features from the RFID data, identify any of the items having feature values associated therewith greater than a threshold amount as inside-cart items to be processed during an automatic checkout, and perform the automatic checkout on only the inside-cart items.

Claims (32)

1. A system for use with a shopping cart in a store, the system comprising:

Radio-Frequency Identification (RFID) tags, including item RFID tags associated with items for sale and reference RFID tags associated with various aisles in the store, for providing RFID data; and

a RFID reader, operatively coupled to the shopping cart, and configured to read RFID data from the item RFID tags and the reference RFID tags, extract features from the RFID data, identify any of the items having feature values associated therewith greater than a threshold amount as inside-cart items to be processed during an automatic checkout, and perform the automatic checkout on only the inside-cart items;

wherein the features comprise a reading count, a median phase value, and a median Doppler shift; and

wherein the RFID reader identifies the inside-cart items using a temporal feature matrix such that each row in the matrix represents a single one of the features measured over time and each column in the matrix represents a set of the features that correspond to a single one of the item RFID tags.

2. The system of claim 1 , wherein the RFID reader identifies the actual inside-cart items by applying a clustering process to the temporal feature matrix to obtain a first cluster of the inside-cart RFID tags and a second cluster of outside-cart RFID tags, wherein the automatic checkout is performed for only the items corresponding to the inside-cart RFID tags.

3. The system of claim 2 , wherein the clustering process is commenced using initial seeds derived from the features.

4. The system of claim 1 , wherein the features further comprise a median Received Signal Strength Indication.

5. The system of claim 1 , wherein the RFID reader is further configured to determine a mobility state of the shopping cart by tracking a variance of a phase and a Received Signal Strength Indication of the RFID tags.

6. The system of claim 1 , wherein the RFID reader is further configured to determine a mobility state of the shopping cart by tracking a population of the RFID tags in an interrogation zone.

7. A method for use with a shopping cart in a store, the method comprising:

reading, by a Radio-Frequency Identification (RFID) reader operatively coupled to the shopping cart, RFID data from Radio-Frequency Identification (RFID) tags including item RFID tags associated with items for sale and reference RFID tags associated with various aisles in the store;

extracting, by the RFID reader, features from the RFID data;

identifying, by the RFID reader, any of the items having feature values associated therewith greater than a threshold amount as inside-cart items to be processed during an automatic checkout; and

performing, by the RFID reader, the automatic checkout on only the inside-cart items;

wherein the features comprise a reading count, a median phase value, and a median Doppler shift; and

wherein the RFID reader identifies the inside-cart items using a temporal feature matrix such that each row in the matrix represents a single one of the features measured over time and each column in the matrix represents a set of the features that correspond to a single one of the item RFID tags.

8. The method of claim 7 , wherein the RFID reader identifies the actual inside-cart items by applying a clustering process to temporal feature matrix to obtain a first cluster of the inside-cart RFID tags and second cluster of outside-cart RFID tags, wherein the automatic checkout is performed for only the items corresponding to the inside-cart RFID tags.

9. The method of claim 8 , wherein the clustering process is commenced using initial seeds derived from the features.

10. The method of claim 7 , wherein the features further comprise a median Received Signal Strength Indication.

11. The method of claim 7 , further comprising determining a mobility state of the shopping cart by tracking a variance of a phase and a Received Signal Strength Indication of the RFID tags.

12. The method of claim 7 , further comprising determining a mobility state of the shopping cart by tracking a population of the RFID tags in an interrogation zone.

13. A non-transitory article of manufacture tangibly embodying a computer readable program which when executed causes a computer to perform the steps of claim 7 .

14. A system for use with a shopping cart in a store, the system comprising:

Radio-Frequency Identification (RFID) tags, including item RFID tags associated with items for sale and reference RFID tags associated with various aisles in the store, for providing RFID data; and

a RFID reader, operatively coupled to the shopping cart, and configured to read RFID data from the item RFID tags and the reference RFID tags, extract features from the RFID data, and identify any of the items having feature values associated therewith that vary greater than a threshold amount over a threshold time duration as items of interest;

wherein the features comprise a reading count, a median phase value, and a median Doppler shift; and

wherein the RFID reader identifies the inside-cart items using a temporal feature matrix such that each row in the matrix represents a single one of the features measured over time and each column in the matrix represents a set of the features that correspond to a single one of the item RFID tags.

15. The system of claim 14 , wherein the RFID reader is further configured to record the feature values of the RFID tags of the items which are in a current one of the various aisles and outside of the shopping cart to identify the items of interest.

16. The system of claim 14 , wherein the RFID reader is further configured to perform filtering of the RFID data, using (i) a precomputed aisle-level filter to reduce an RFID interrogation zone of the reader in a current one of the various aisles and (ii) an item-level filter to filter out the inside-cart items from the outside-cart items to identify the actual inside-cart items.

17. The system of claim 14 , further comprising a display device operatively coupled to the shopping cart, wherein the RFID reader causes an action to be performed on the display device relative to at least one of the items of interest to enhance an interest of the user in the at least one of the items of interest.

18. The system of claim 14 , wherein the feature values that vary greater than the threshold amount over the threshold time duration are determined using Kullback-Leibler divergence.

Assignments (2)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jun 11, 2019
From: NEC LABORATORIES AMERICA, INC.
To: NEC CORPORATION
Reel/Frame 049427/0718 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Apr 16, 2018
From: SUNDARESAN, KARTHIKEYAN; CHAI, EUGENE; RANGARAJAN, SAMPATH
To: NEC LABORATORIES AMERICA, INC.
Reel/Frame 045554/0014 →