Basket-Linked Commodity Verification for Self-Checkout Fraud
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional fraud prevention systems for self-checkout machines require complex data management and installation of new equipment, incurring high costs and operational burdens on stores.
Innovation Solution
A fraud detection system that uses image acquisition and analysis to associate commodity-picking images with identification information, allowing for real-time matching of registered and confirmed commodities, and alerts store clerks to mismatches.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional fraud prevention systems use in-store cameras and smart carts to monitor customer behavior and manage commodity information, then fraud detection capability is improved, but device complexity and installation cost increase
Solution Approach 1:
The patent merges the fraud detection function with the existing self-checkout machine, eliminating the need for separate camera systems and smart cart infrastructure. The self-checkout machine integrates image acquisition, commodity specification, and fraud detection capabilities into a single unified system, reducing overall device complexity while maintaining detection reliability.
Solution Approach 2:
The self-checkout machine is designed to perform multiple functions: it serves as both the primary checkout device for customers and the fraud detection system. By making the self-checkout machine universal, the patent eliminates the need for dedicated fraud prevention equipment, thereby reducing installation costs and system complexity while preserving fraud detection capability.
2Measurement precision
If conventional systems capture and manage detailed customer behavior data and commodity information, then fraud detection accuracy is improved, but data management complexity and operational burden increase
Solution Approach 1:
The system automatically performs fraud detection without requiring manual data management or intervention. The self-checkout machine autonomously captures images, specifies commodities using machine learning, compares registered versus actual commodities, and generates alerts. This self-service approach maintains high detection accuracy while eliminating the operational burden of manual data handling.
Solution Approach 2:
The system implements automatic feedback loops where the self-checkout machine continuously monitors the checkout process, compares expected commodities with actual scanned items, and immediately alerts staff to discrepancies. This real-time feedback mechanism ensures accurate fraud detection without requiring manual data verification, thereby reducing operational burden.
3Reliability
If conventional fraud prevention systems install new machines and equipment in stores, then fraud detection capability is improved, but installation cost increases
Solution Approach 1:
The patent combines fraud detection functionality with the existing self-checkout machine infrastructure. By merging these functions, the system eliminates the need to install separate camera systems, smart cart technology, or dedicated fraud prevention equipment, thereby significantly reducing installation costs while maintaining effective fraud detection capability.
Solution Approach 2:
The self-checkout machine is designed as a universal device that performs both customer checkout operations and fraud detection. This multi-functionality eliminates the need for additional specialized equipment, reducing capital expenditure and installation costs for stores while ensuring reliable fraud detection through integrated image acquisition and analysis capabilities.
Data Source
AI summary
A camera captures an image of a customer picking a commodity, and transmits the image to a management device. The management device specifies the picked commodity from the received image, and stores the same into purchase-planned commodity data in association with a basket ID assigned to each basket. When the customer has performed checkout with a self-checkout machine, the self-checkout machine transmits data of a settled commodity to the management device. The management device stores the received data of the commodity into checkout commodity data in association with the basket ID, and judges matching between the purchase-planned commodity and the checkout commodity. When the result of the judgement is “mismatch”, a clerk is notified of this mismatch through a clerk terminal.