Self-Checkout Fraud Detection Using Behavior Tracking and Payment Status
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing fraud detection systems in self-service checkouts often incorrectly identify individuals who do not need to pay as attempting to bypass checkout, leading to excessive detection and increased monitoring burdens.
Innovation Solution
A fraud detection system that tracks individuals' behaviors within a payment area, calculates a determination score based on payment-required actions, and assigns payment completion information, determining fraudulent passage based on these criteria.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If all people passing through exits without payment are detected as checkout-bypassing targets, then fraud detection coverage is improved, but excessive detection occurs for people who do not require payment
Solution Approach 1:
The patent applies local quality by differentiating detection criteria for different people based on their characteristics. Companions and clerks are identified as special groups that should not be detected as fraud, while regular shoppers are subject to normal detection. This localized differentiation allows the system to maintain high detection coverage for actual fraud while avoiding false positives for legitimate non-paying individuals.
Solution Approach 2:
The patent segments the population into different categories (companions, clerks, regular shoppers) and applies different detection rules to each segment. By dividing the detection target into specific groups with distinct characteristics, the system can accurately identify fraud without mistakenly flagging people who have legitimate reasons for not paying at the checkout.
2Measurement precision
If visual monitoring by clerks is implemented to prevent excessive detection, then detection accuracy is improved, but monitoring burden on clerks increases
Solution Approach 1:
The patent implements self-service by enabling the detection system to automatically identify and exclude companions and clerks from fraud detection without requiring manual intervention. The system autonomously analyzes video data to recognize these special groups and adjusts detection accordingly, eliminating the need for continuous visual monitoring by clerks while maintaining high detection accuracy.
Solution Approach 2:
The patent replaces the mechanical visual monitoring system with an automated image processing system. Instead of relying on clerks to visually identify and monitor potential fraud, the system uses computer vision algorithms to automatically detect checkout-bypassing behavior, reducing human burden while improving consistency and accuracy.
3Productivity
If automated payment systems are implemented, then payment efficiency is improved, but fraudulent passage detection becomes more challenging
Solution Approach 1:
The patent introduces video capture devices and image processing algorithms as intermediaries between the automated payment system and fraud detection. These intermediaries monitor the checkout area and exit points, capturing visual evidence of fraudulent passage while allowing the automated payment system to continue operating efficiently. The intermediary layer enables fraud detection without disrupting the streamlined payment process.
Solution Approach 2:
The patent adds a visual monitoring dimension to the automated payment system. By incorporating video capture and image analysis alongside the existing payment processing, the system creates a multi-dimensional approach that maintains payment efficiency while enabling fraud detection through a separate observational channel.
Data Source
Figure 1
Figure 2
Figure 3
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.