AI Cart Image Verification for Self-Checkout Scanning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Customers struggle to remember which items have been scanned when using a handheld scanner in a shopping cart due to heavy or irregular items, leading to inefficiencies and retail shrinkage in self-checkout transactions.
Innovation Solution
Implementing an AI/ML-based scanning verification system that uses cameras to capture images of items in the shopping cart, processes them to identify scanned items, and displays a visual indication on the self-checkout system to ensure all items are scanned, preventing incomplete transactions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If customers use handheld scanners to scan items in shopping carts without removing items, then scanning convenience is improved, but the ability to remember scanned items deteriorates
Solution Approach 1:
The system provides real-time visual feedback by displaying scanned items on a monitor, allowing customers to see which items have been scanned and which remain unscanned. This feedback mechanism eliminates the need for customers to remember scanned items manually.
Solution Approach 2:
The system creates a visual copy representation of scanned items on the display screen, replacing the customer's memory function. The visual display serves as an external memory device that tracks and presents scanning status.
2Reliability
If customers manually track scanned items, then scanning accuracy is improved, but transaction time increases
Solution Approach 1:
The system performs self-tracking of scanned items automatically through image recognition and processing, eliminating the need for customers to manually monitor and track scanned items. The system serves itself by automatically maintaining the scanning status record.
Solution Approach 2:
The patent replaces manual mechanical tracking (customer remembering and noting scanned items) with an automated optical recognition system using cameras and image processing algorithms to track scanning status.
3Reliability
If the system displays all items in the shopping cart, then scanning verification is improved, but device complexity increases
Solution Approach 1:
The system uses a single camera to perform multiple functions: capturing images of items in the shopping cart, recognizing item identities, tracking scanning status, and providing visual feedback. This multi-functionality reduces the need for multiple separate devices.
Solution Approach 2:
The patent introduces an intermediary image processing system that bridges the camera and the scanning verification function. The image processing algorithms act as intermediaries to translate visual data into scanning status information.
Data Source
AI summary
Methods and apparatus for performing artificial intelligence (AI)-based scanning verification for a self-checkout transaction using a self-checkout system are described. In an example method, a first image of multiple items within a shopping cart is displayed on a display system of the self-checkout system. A scanning of a first item of the multiple items with a handheld scanner associated with the self-checkout system is detected. A second image of at least a second item of the multiple items within the shopping cart is generated. The second image lacks an indication of the first item. The second image is displayed on the display system.