Vision-based frictionless self-checkouts for small baskets
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Convenience stores face challenges with small baskets and lack of space for self-service terminals, leading to busy queues and potential loss of sales due to shoppers abandoning purchases, while existing image recognition technologies are not optimized for small transactions.
Innovation Solution
A vision-based self-checkout system using three cameras and a base with integrated scale to capture and process images from multiple angles, enabling accurate item recognition and pricing without the need for barcodes, suitable for compact spaces.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If traditional barcode scanning systems are used, then item identification can be achieved, but the system requires physical space for Self-Service Terminals which is not available in convenience stores
Solution Approach 1:
The patent replaces traditional mechanical barcode scanning systems with a vision-based system using cameras and image processing algorithms. The system captures images of items, processes them through neural networks to identify products, and determines pricing without requiring physical barcode labels or traditional scanning hardware, thereby eliminating the need for large Self-Service Terminals.
Solution Approach 2:
The vision-based system serves multiple functions: it identifies items, determines pricing, and processes transactions all through image analysis. The same camera system that captures item images also enables the entire checkout process, combining multiple functions into a single compact unit that doesn't require dedicated terminal space.
2Measurement precision
If store assistants are used to scan items, then accurate item identification is achieved, but queue congestion increases and transaction time is extended
Solution Approach 1:
The system enables customers to perform their own checkouts automatically through image capture and processing. The vision system autonomously identifies items, looks up pricing, and processes transactions without requiring store assistant intervention, allowing customers to complete purchases independently and rapidly.
Solution Approach 2:
The image processing and item identification occur continuously and automatically as items are placed in the basket or on the scanning surface. The system processes multiple items in sequence without interruption, maintaining continuous transaction flow rather than requiring sequential manual scanning operations.
3Measurement precision
If manual item scanning is used, then item codes can be entered accurately, but the process becomes inconvenient and shoppers may abandon purchases
Solution Approach 1:
The system replaces manual barcode scanning and code entry with automated vision-based item identification. Cameras capture images of items, and image processing algorithms automatically identify products and retrieve pricing information, eliminating the need for customers to manually scan or enter codes.
Solution Approach 2:
The system creates digital copies of items through image capture and uses these visual copies for identification and pricing lookup. Instead of requiring physical barcode labels, the system processes visual information from captured images to identify items and determine costs.
Data Source
Figure 1A
Figure 1B
Figure 1C
AI summary
A vison-based self-checkout terminal is provided. Purchased items are placed on a base and multiple cameras take multiple images of each item placed on the base. A location for each item placed on the base is determined along with a depth and the dimensions of each item at its given location on the base. Each item's images are then cropped, and item recognition is performed for each item on that item's cropped images with that item's corresponding depth and dimension attributes. An item identifier for each item is obtained along with a corresponding price and a transaction associated with items are completed.