On-Shelf Barcode Reader Using Low-Resolution Cameras
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Inventory management systems in retail stores and warehouses face inefficiencies in tracking millions of products due to inaccurate scanning methods, leading to inventory discrepancies and increased labor costs, especially during high traffic periods, which can result in stockouts and delays.
Innovation Solution
An inventory visibility management system using low-resolution cameras to capture images, process them locally or in the cloud, and apply machine-learned models to generate high-resolution images of product identifiers, enabling accurate tracking and reducing the need for manual scanning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If high-resolution cameras are used to capture product identifiers, then measurement precision is improved, but device complexity and cost increase
Solution Approach 1:
The system divides the image processing task into two stages: first, low-resolution cameras capture multiple images of products; second, a super-resolution algorithm processes these multiple low-resolution images to generate a high-resolution composite image of the identifier. This segmentation allows the use of simple cameras while achieving high measurement precision through computational processing.
Solution Approach 2:
The patent introduces a super-resolution algorithm as an intermediary processing step between image capture and identifier recognition. This intermediary takes multiple low-resolution images as input and produces a high-resolution output image, effectively bridging the gap between simple camera hardware and high-precision identifier reading requirements.
2Measurement precision
If multiple low-resolution images are processed to generate high-resolution images, then measurement precision is improved, but use of energy and computational resources increase
Solution Approach 1:
The system extracts only the identifier region from the full product images using object detection and cropping techniques. By focusing computational resources on processing only the relevant identifier portions rather than entire images, the energy consumption is significantly reduced while maintaining high measurement precision for the critical identifier recognition task.
Solution Approach 2:
The patent captures more images than strictly necessary (excessive action) to provide sufficient input data for the super-resolution algorithm, but processes only the essential identifier regions (partial action) rather than entire images. This approach ensures high measurement precision while controlling computational energy usage by being selective about what gets processed.
Data Source
AI summary
An inventory visibility management system utilizes fixed or motorized cameras to scan inventory bearing shelves in a backroom or warehouse of a store, as opposed to store shelves where merchandise is available for purchase, for inventory frequently to keep the system up to date on what boxes of inventory are on the shelf, what is in those boxes and where those boxes are on the shelf. The system may identify a bounding polygon around an identifier corresponding to the product and apply the bounding polygon to a machine-learned model, which may generate a high-resolution crop of the identifier as output. The system registers the identifier to the first bounding polygon and to a location associated with cameras that captured the plurality of low-resolution images. Upon receiving a request from a client device, the system may provide the location associated with the one or more cameras to the client device.