Imaging used to reconcile cart weight discrepancy
Disclosed embodiments provide transaction validation systems and methods for use in self-checkout at a retail establishment. An exemplary system may include a scale sized to support and weigh a shopping basket thereon. At least one image sensor is arranged to capture at least one image of the weighed shopping basket. A processor receives a weight measurement of the shopping basket from the scale and determines, based on information received from a wireless communications device of a customer, a record of products in the shopping basket and an expected cumulative weight associated with the products in the shopping basket. The processor determines a discrepancy between the weight measurement of the products in the shopping basket and the expected cumulative weight and reconciles the determined discrepancy using the at least one captured image of the shopping basket by the at least one image sensor.
1. A transaction validation system for use in self-checkout at a check-out location, comprising:
a scale sized to support and weigh a plurality of items, and to track weight changes as one or more items are added to or removed from said scale;
one or more image sensors to acquire an image of each item placed on said scale; and
at least one processor configured to:
receive an updated weight measurement of all items placed on said scale based on said weight change tracking,
identify each item placed on said scale based on said acquired image,
determine an expected cumulative weight of all items placed on said scale based on said acquired image for each item, wherein said expected cumulative weight takes into account an expected deviation in a weight of each item, and
compare said expected cumulative weight with said updated weight measurement.
2. The transaction validation system of claim 1 , wherein said at least one processor is further configured to distinguish between items of a different size and of a same type based on said acquired image and said updated weight measurement.
3. The transaction validation system of claim 1 , wherein said weight of each item and said expected deviation in said weight of said item is an average weight of said item and a standard deviation based on weights of all items of a same type.
4. The transaction validation system of claim 1 , wherein said at least one processor is further configured to determine a discrepancy between said expected cumulative weight and said updated weight measurement when a difference between said expected cumulative weight and said updated weight measurement exceeds a statistical threshold.
5. The transaction validation system of claim 4 , wherein said statistical threshold is associated with a confidence level established by a retail establishment associated with the check-out location.
6. The transaction validation system of claim 4 , wherein said statistical threshold is associated with a confidence level based on a trust level associated with a customer making a transaction in a retail establishment.
7. The transaction validation system of claim 6 , wherein said trust level is determined by a number, frequency, and/or size of transactions between said customer and said retail establishment.
8. The transaction validation system of claim 4 , wherein said at least one processor is further configured to recommend reconciling said discrepancy by adding or removing an item, or by replacing an item.
9. The transaction validation system of claim 1 , wherein said plurality of items is contained inside a shopping basket.
10. A method of validating a transaction in a self-checkout system at a check-out location, the method comprising:
weighing a plurality of items on a scale and tracking weight changes as one or more items are added to or removed from said scale;
acquiring an image of each item placed on said scale with one or more image sensors; and,
using at least one processor:
receiving an updated weight measurement of all items placed on said scale based on said weight change tracking,
identifying each item placed on said scale based on said acquired image,
determining an expected cumulative weight of all items placed on said scale based on said acquired image for each item, wherein said expected cumulative weight takes into account an expected deviation in a weight of each item, and
comparing said expected cumulative weight with said updated weight measurement.
11. The method of claim 10 , further comprising distinguishing between items of a different size and of a same type based on said acquired image and said updated weight measurement.
12. The method of claim 10 , wherein said weight of each item and said expected deviation in said weight of said item is an average weight of said item and a standard deviation based on weights of all items of a same type.
13. The method of claim 10 , further comprising determining a discrepancy between said expected cumulative weight and said updated weight measurement when a difference between said expected cumulative weight and said updated weight measurement exceeds a statistical threshold.
14. The method of claim 13 , wherein said statistical threshold is associated with a confidence level established by a retail establishment associated with the check-out location.
15. The method of claim 13 , wherein said statistical threshold is associated with a confidence level based on a trust level associated with a customer making a transaction in a retail establishment.
16. The method of claim 15 , wherein said trust level is determined by a number, frequency, and/or size of transactions between said customer and said retail establishment.
17. The method of claim 13 , further comprising recommending reconciling said discrepancy by adding or removing an item, or by replacing an item.
18. The method of claim 10 , wherein said plurality of items is contained inside a shopping basket.
19. A non-transitory computer readable medium comprising instructions for use at a self-checkout at a retail establishment, the computer readable medium containing instructions that when executed by at least one processor cause the at least one processor to perform the following:
receive an updated weight measurement of all items placed on a scale based on weight change tracking as one or more items are added to or removed from said scale,
identify each item placed on said scale based on an acquired image of each item,
determine an expected cumulative weight of all items placed on said scale based on said acquired image for each item, wherein said expected cumulative weight takes into account an expected deviation in a weight of each item, and
compare said expected cumulative weight with said updated weight measurement.
20. The non-transitory computer readable medium of claim 19 , wherein said weight of each item and said expected deviation in the weight of said item is an average weight of the item and a standard deviation based on weights of all items of a same type.