Hough Voting for SKU Size Differentiation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image recognition techniques fail to distinguish between products that have similar appearances but vary in size, making it difficult to automate the enforcement of planograms in retail environments.
Innovation Solution
A system and method that uses an image recognition application to identify products, generate regions of interest, determine pixel dimensions, calculate pixel-to-physical dimension ratios, and assign stock keeping unit identifiers based on these ratios, allowing for accurate differentiation of products with similar packaging but varying sizes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional image recognition techniques are used to identify products, then product identification can be performed, but products with similar appearances but different sizes cannot be distinguished
Solution Approach 1:
The patent applies parameter changes by utilizing pixel dimension measurements as an additional distinguishing parameter beyond visual appearance. By measuring the pixel dimensions of products in the image and comparing them against known physical dimensions, the system can differentiate between products of similar appearance but different sizes, thereby resolving the contradiction between identification reliability and size differentiation precision.
2Measurement precision
If manual tracking of product location and quantity is performed, then accurate inventory data can be obtained, but the process is labor-intensive and difficult to enforce planograms
Solution Approach 1:
The patent replaces the manual mechanical tracking system with an automated image recognition and measurement system. By capturing images of retail shelves and automatically analyzing product locations, quantities, and dimensions through computational methods, the system eliminates labor-intensive manual tracking while maintaining accurate inventory data and enabling efficient planogram enforcement.
Solution Approach 2:
The system enables self-service by allowing the image recognition algorithm to automatically perform inventory tracking and planogram compliance checking without human intervention. The automated measurement and identification processes eliminate the need for manual data collection, thereby improving productivity while maintaining measurement precision.
3Productivity
If automated product recognition is attempted, then productivity increases, but the ability to distinguish between similar products varies in size is lost
Solution Approach 1:
The patent adds another dimension to product recognition by incorporating pixel dimension measurements alongside visual appearance analysis. This dimensional extension allows the automated system to differentiate between products of similar appearance by measuring their sizes in the image, thereby maintaining both high productivity and accurate size-based differentiation.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The disclosure includes a system and method for distinguishing between items of similar appearance that vary in size. An image recognition application receives an image including a plurality of items, identifies each item in the image, generates a region of interest for each identified item in the image, determines a dimension of the region of interest for each identified item in pixels, determines a plurality of pixel-to-physical dimension ratios for each identified item, identifies a selected pixel-to-physical dimension ratio among the plurality of pixel-to-physical dimension ratios, and determines a stock keeping unit identifier of each identified item in the image based on the selected pixel-to-physical dimension ratio and the dimension of the region of interest for each identified item.