Computer Vision Basket Matching for Real-Time Exit Verification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing retail checkout systems incur inefficiencies and customer inconvenience due to manual scanning of shopping carts at exit to verify unpaid items, leading to potential financial loss and negative customer experience.
Innovation Solution
A computer vision-based system that uses image capture devices to recognize items in shopping carts and matches them with candidate receipts using confidence scores, eliminating the need for manual scanning by comparing item recognition results with candidate receipts and assigning weights based on uniqueness and frequency of items.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual scanning of shopping carts is performed at store exit to verify unpaid items, then accuracy in identifying unscanned items is improved, but customer experience deteriorates and exit time increases
Solution Approach 1:
The system performs preliminary scanning of items as customers leave the store, automatically capturing images of shopping carts and comparing them against receipt data before customers complete their exit. This eliminates the need for customers to stop for manual verification, maintaining both high accuracy in identifying unscanned items and fast exit times.
Solution Approach 2:
The patent replaces the manual mechanical scanning process with an automated computer vision system using cameras and image processing algorithms. The system automatically captures images of items in shopping carts, identifies products, and compares them with receipt data, eliminating the need for manual intervention and significantly reducing exit time while maintaining measurement precision.
2Reliability
If manual scanning procedures are implemented at store exit, then verification of paid items is improved, but operational efficiency deteriorates
Solution Approach 1:
The system enables self-service verification where the automated computer vision system independently performs the entire verification process - capturing images, identifying items, comparing with receipts, and generating alerts for unscanned items. This eliminates the need for employee intervention, maintaining reliable verification while dramatically improving operational efficiency by freeing staff from manual scanning tasks.
Solution Approach 2:
The patent replaces manual verification operations with an automated image recognition system that processes shopping cart contents, matches items against receipts, and identifies discrepancies without human intervention. This substitution maintains verification reliability while improving productivity by enabling the system to handle multiple carts simultaneously and reducing the time required per verification.
3Speed
If computer vision-based automatic matching is implemented, then processing speed is improved, but system complexity increases
Solution Approach 1:
The patent divides the complex verification system into modular functional components: image capture modules, image processing modules, item identification modules, receipt comparison modules, and alert generation modules. Each module performs a specific function independently, which maintains processing speed while making the overall system easier to manage, maintain, and scale compared to a monolithic system.
Data Source
AI summary
Examples provide a system for exit computer vision basket matching receipts to shopping carts automatically in real-time as a customer is exiting a retail facility. Computer vision analysis of one or more images of a shopping cart produces a set of recognized items from the images. The set of recognized items are compared to the items identified in each candidate receipt in a plurality of candidate receipts. A confidence score is generated for each candidate receipt based on the number of common items found in both the set of recognized items and each receipt. The score is weighted based on factors such as uniqueness of the items and number of instances of each item. The receipt having the highest weighted confidence score is paired with the shopping cart. The paired receipt and image of the shopping cart are output to a user for verification via a user interface.