| CPC G06Q 20/389 (2013.01) [G06N 20/00 (2019.01); G06Q 20/401 (2013.01); G06Q 30/0633 (2013.01); G06Q 30/0641 (2013.01); G06T 7/00 (2013.01); G06V 30/18 (2022.01)] | 19 Claims |
|
1. A method, comprising:
obtaining an image of a transaction area during a transaction at a transaction terminal based on a transaction event and using metadata comprising a camera identifier, a time stamp, and location information associated with the image to identify a specific transaction terminal for the transaction terminal where the transaction event occurred, and a specific store associated with the transaction terminal;
wherein obtaining further includes:
obtaining the image as a last image captured of the transaction area just before a payment request is made to complete the transaction at the transaction terminal;
determining whether a cart/basket is present within the image;
determining whether the cart/basket is nonempty or empty when the cart/basket is present within the image using a machine learning model that processes the image to detect a presence of the cart/basket based on learned characteristics of carts/baskets specific to an environment of the specific store;
wherein the machine learning model is trained on images of transaction areas to recognize and identify, within the images, boundaries and a particular transaction area associated with a cart/basket;
wherein during re-training to account for false cart/basket detections or no cart/basket detection, each original trained image is replicated, rotated, and manipulated for scaling and/or brightness variations and used to re-train the machine learning model;
determining whether the cart/basket includes at least one saleable item when the cart/basket is nonempty; and
causing the transaction to suspend for intervention of the transaction when the cart/basket includes the at least one saleable item.
|