Self-Checkout Fraud Detection Using Product Path and Dwell Time

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing fraud detection systems at self-checkout terminals struggle to accurately identify fraudulent actions, such as scan skipping, using only captured images from a camera without cooperation with the self-checkout terminal or POS system.

Innovation Solution

A fraud detection system that analyzes the movement path and residence time of products in multiple image areas using a camera, determining fraudulent actions by recognizing products from captured images and detecting scan skipping based on both movement path and speed, without requiring cooperation with the self-checkout terminal or POS system.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If fraud detection is performed using only captured images from a camera without cooperation with self-checkout terminal or POS system, then ease of operation and versatility are improved, but measurement precision of fraudulent actions deteriorates

Engineering Contradiction:
Improveease of installationVSAvoiddetection accuracy
Core Design Contradiction:
Ease of operationVSMeasurement precision

Solution Approach 1:

The monitoring area is divided into multiple image areas (first image area closest to scanner, second image area, etc.) to track product movement through different zones. This segmentation enables precise measurement of movement path and residence time using only camera images, resolving the contradiction between independent operation and detection accuracy.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system adds temporal dimension by measuring residence time of products in each image area, and spatial dimension by tracking movement paths across multiple areas. This multi-dimensional analysis from 2D camera images enables accurate fraud detection without requiring integration with terminal systems.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Measurement precision

If multiple image areas and residence time analysis are used to detect fraudulent actions, then measurement precision of scanning operations is improved, but device complexity increases

Engineering Contradiction:
Improvedetection accuracyVSAvoidsystem complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system uses the camera's existing image capture capability to perform fraud detection by itself, without requiring additional sensors or cooperation from the self-checkout terminal. The image processing unit analyzes movement paths and residence times directly from captured images, enabling the system to serve itself for detection purposes.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The system changes the analysis parameters from simple presence detection to movement path tracking and residence time measurement. By analyzing how long products remain in each image area and their movement trajectories, the system achieves high detection accuracy using only standard camera equipment.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP4715712A1Computer program, fraud detection apparatus, and fraud detection system
Publication Date: 2026.03.25 FUJITSU LTD
  • EP4715712A1 patent drawingFigure 1
  • EP4715712A1 patent drawingFigure 2
  • EP4715712A1 patent drawingFigure 3

AI summary

A storing unit stores area information indicating positions of a plurality of image areas set in a captured image of a front region of a self-checkout terminal including a scanner. A processing unit recognizes a product from the captured image. The processing unit detects a fraudulent action related to a scanning operation for causing the scanner to scan product information attached to the product, based on a movement path of the recognized product in the plurality of image areas, and a residence time of the product in a first image area that is closest to the scanner among the plurality of image areas.