Self-Checkout Fraud Detection Using Behavior Scoring and Payment Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The implementation of fully self-service checkouts is plagued by checkout-bypassing fraud, where customers take items without paying, placing a heavy burden on clerks due to visual monitoring inefficiencies.
Innovation Solution
A fraud detection system that tracks individuals within a payment area, calculates a determination score based on payment-required behaviors, and determines fraudulent passage by analyzing behavior patterns to prevent excessive detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If visual monitoring by clerks is implemented to detect checkout-bypassing, then fraud detection capability is improved, but labor burden and operational complexity increase
Solution Approach 1:
The patent replaces the mechanical visual monitoring system operated by clerks with an automated image processing system using cameras and algorithms. The system captures images, detects objects and people, tracks their movements, and automatically determines whether checkout-bypassing has occurred, eliminating the need for manual visual monitoring while maintaining or improving detection reliability.
2Device complexity
If automated image processing is used to detect checkout-bypassing, then labor burden is reduced, but false alarm rate increases
Solution Approach 1:
The patent segments the detection process into distinct functional modules: object detection, people detection, tracking, payment status determination, and fraudulent passage determination. Each module handles a specific aspect of the analysis, allowing for more precise and nuanced judgment. The system separately tracks objects and people, determines payment status independently, and then combines this information to make final fraud determination, reducing false alarms through structured multi-stage analysis.
Solution Approach 2:
The patent changes the parameters used for fraud detection from simple presence/absence detection to multi-dimensional analysis including tracking trajectories, determining payment status, and calculating determination scores. By introducing these additional parameters and using composite scoring rather than binary detection, the system achieves more accurate fraud identification with fewer false alarms.
3Measurement precision
If comprehensive tracking and behavior analysis is implemented, then detection accuracy is improved, but system complexity and processing time increase
Solution Approach 1:
The patent divides the complex tracking and analysis system into separate functional units: an object detection unit, a people detection unit, a tracking unit, and a determination unit. Each unit performs a specific function independently, which simplifies the overall system architecture while enabling comprehensive analysis. This modular segmentation allows the system to maintain high detection accuracy without excessive complexity.
4Reliability
If behavior pattern analysis is used to calculate determination scores, then false alarm reduction is achieved, but processing time and computational load increase
Solution Approach 1:
The patent performs preliminary tracking and behavior analysis continuously as objects and people move through the store, calculating determination scores in real-time rather than analyzing all data at once. The system pre-processes movement trajectories and payment status information during the shopping process, so that when fraud determination is needed at the exit, the analysis is already complete or near-complete, reducing final processing time while maintaining accurate false alarm reduction.
Data Source
AI summary
A fraud detection device includes a processor that executes a procedure. The procedure includes: detecting and tracking people who are within a predetermined range based on sensor data acquired through sensing in the predetermined range in a store including a payment area, where a self-service checkout is installed, and recognizing a behavior of each of the tracked people; calculating a determination score indicating an extent to which each of the tracked people is a person who is required to perform payment at the self-service checkout, based on the recognized behavior; assigning payment information, indicating that payment is completed, to a person who paid at the self-service checkout; and determining fraudulent passage at an exit of the store for each person passing through the exit, based on the determination score and the payment information.