Time-Stamped Image Recognition for Frictionless Checkout
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current shopping checkout processes require customers to handle items multiple times for scanning, leading to customer frustration and queue growth, especially during peak times, as they need to pick, scan, bag, and remove items from bags, and self-checkout systems can be cumbersome with barcode scanning requirements.
Innovation Solution
A system utilizing cameras and machine-learning algorithms to capture and process time-stamped images of items and people, allowing for item recognition without manual scanning, where cropped images are input into a trained machine-learning item identifier to determine item identities with confidence values, and these are used to manage transactions efficiently.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If manual barcode scanning is used at checkout terminals, then item identification can be performed, but customers must handle items multiple times and the process becomes cumbersome
Solution Approach 1:
The patent replaces the mechanical barcode scanning system with an image recognition system using cameras and machine learning algorithms. The system captures images of items on shelves and automatically identifies them, eliminating the need for customers to manually scan barcodes and handle items multiple times for identification purposes.
Solution Approach 2:
The system enables self-service by automatically tracking items as customers pick them up from shelves. The image recognition system continuously monitors the store environment and automatically identifies items without requiring customer intervention for scanning or manual identification, allowing customers to simply walk away with items.
2Productivity
If self-service checkout terminals are used, then checkout speed can be improved, but customers experience frustration and queues grow during peak times
Solution Approach 1:
The patent implements a comprehensive self-service system where customers can shop and checkout without staff assistance. The image recognition system automatically tracks items in carts and identifies them as customers move through the store, eliminating the need for customers to manually scan items or navigate touchscreen interfaces during checkout.
Solution Approach 2:
The system performs preliminary item identification in advance as customers shop, before they reach the checkout area. By continuously monitoring and pre-identifying items through image recognition throughout the store, the system eliminates the need for customers to locate barcodes or navigate interfaces at checkout, reducing friction during peak times.
3Measurement precision
If traditional checkout terminals with scanners are used, then item identification can be performed, but dedicated terminals and scanning hardware are required
Solution Approach 1:
The patent makes the imaging system universal by using standard store cameras to perform multiple functions: monitoring customer movement, tracking items in carts, and identifying products. This eliminates the need for dedicated scanning hardware at terminals, as the same camera infrastructure serves all checkout and inventory tracking needs throughout the store.
Solution Approach 2:
The system replaces dedicated mechanical barcode scanners with an optical image recognition system. Instead of requiring specialized scanning hardware at checkout terminals, the patent uses general-purpose cameras and machine learning algorithms to identify items, reducing hardware complexity while maintaining or improving identification capabilities.
Data Source
AI summary
Time-stamped images are captured for unknown items within a store. Portions of the images are cropped to restrict the cropped images to attributes associated with the unknown items. A set of time ordered cropped images are proved to a trained machine-learning item recognition algorithm, which produces an output of confidence values that map to specific items of the store. When the confidence value meets or exceeds a predefined threshold, the specific item associated with that confidence value is used to identify the unknown item. The known item is assigned to an appropriate shopping cart of a shopper based on tracking from the images of the shopper while in the store.