Cart inspection for suspicious items
Methods and apparatus provide for a Cart Inspector to create a suspicion level for a transaction when a video image of the transaction portrays an item(s) left in a shopping cart. Specifically, the Cart Inspector obtains video data associated with a time(s) of interest. The video data originates from a video camera that monitors a transaction area. The Cart Inspector analyzes the video data with respect to target image(s) associated with a transaction in the transaction area during the time(s) of interest. The Cart Inspector creates an indication of a suspicion level for the transaction based on analysis of the target image(s). Creation of a high suspicion level for the transaction indicates that the transaction's corresponding video images most likely portray occurrences where the purchase price of an item transported through the transaction area was not included in the total amount paid by the customer.
1. A method to identify suspicious activity, the method comprising:
in a system comprising a computer device linked to a video camera, the computer device comprising a least a processor and a memory, the memory comprising at least an operating system and a cart inspector process, utilizing video captured by the video camera to track a shopping cart in a transaction checkout area;
sending the video to the cart inspector process;
in the cart inspector process, comparing imagery of the shopping cart in the video with a reference image of an empty cart to determine a degree to which the shopping cart captured by the video is empty;
in the cart inspector process, assigning a suspicion level to the shopping cart based on the degree of the shopping cart being empty; and
in response to detecting that the suspicion level is above a threshold value, in the cart inspector process, generating a notification.
2. The method as in claim 1 , wherein comparing the imagery of the shopping cart in the video with the reference image of the empty cart further comprises:
in the cart inspector process, calculating a set of parameters from the imagery of the shopping cart; and
in the cart inspector process, comparing the set of parameters with model parameters of the shopping cart.
3. The method as in claim 1 , wherein comparing the imagery of the shopping cart in the video with the reference image of the empty cart to determine the degree to which the shopping cart is empty further comprises:
comparing pixels of the imagery of the shopping cart in the video with corresponding pixels in the reference image of the empty cart to determine the degree to which the shopping cart is empty.
4. The method as in claim 1 further comprising:
in the cart inspector process, identifying a region in the imagery of the shopping cart, the identified region being representative of an object;
in the cart inspector process, classifying the object as a non-retail item; and
in the cart inspector process, adjusting the suspicion level in response to detecting that the object is a non-retail item.
5. The method as in claim 4 further comprising:
in the cart inspector process, classifying the object in the shopping cart as being a non-retail item based on detecting motion associated with the object.
6. The method as in claim 4 , wherein classifying the object includes detecting a non-empty region within a location of the shopping cart.