Pixel-to-Physical Ratio for SKU Size Differentiation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image recognition techniques fail to distinguish between objects that are similar in appearance but vary in size, making it difficult to automatically recognize and manage products in retail environments effectively.
Innovation Solution
A system and method that uses an image recognition application to identify items in an image, generate a region of interest for each item, determine pixel dimensions, and calculate a pixel-to-physical dimension ratio using a reference marker to accurately identify and distinguish between stock keeping units of varying sizes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If existing image recognition techniques are used to identify products, then product identification can be performed automatically, but the system fails to distinguish between products of similar appearance but different sizes
Solution Approach 1:
A reference marker with known physical dimensions is introduced as an intermediary object in the image. This marker serves as a mediator between the image recognition system and the products to be measured, providing a basis for calculating real-world dimensions from pixel measurements and enabling accurate size differentiation of products with similar appearances
Solution Approach 2:
The system changes the parameter of measurement by transitioning from purely visual appearance-based recognition to dimension-based recognition. By calculating the ratio between pixel dimensions and actual physical dimensions using the reference marker, the system can differentiate products based on their size parameters even when their visual appearance is similar
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 mechanical manual tracking system with an automated image recognition and measurement system. By using computer vision to automatically identify products, calculate their dimensions through reference marker comparison, and determine inventory status, the system eliminates labor-intensive manual counting while maintaining accurate inventory data
Solution Approach 2:
The system enables self-service inventory management by automatically performing measurements and calculations without human intervention. The image recognition system independently identifies products, determines their sizes through pixel-to-physical dimension ratio calculations, and provides inventory data, freeing staff from manual tracking tasks
3Measurement precision
If a reference marker with known physical dimensions is introduced, then accurate size measurement can be achieved, but the system complexity increases
Solution Approach 1:
The reference marker acts as a simple intermediary that bridges the gap between pixel measurements and real-world dimensions. Rather than implementing complex calibration procedures or multiple measurement devices, the system uses a single marker with known dimensions to establish a conversion ratio, simplifying the overall measurement system while maintaining accuracy
Data Source
AI summary
The disclosure includes a system and method for distinguishing between stock keeping units of similar appearance that vary in size. An image recognition application receives an image depicting a plurality of items, the image including a reference marker with a known physical dimension. The image recognition application performs image recognition to identify an item in the image and a region of interest for the identified image. The image recognition application further determines a pixel-to-physical dimension ratio using the dimension of a region of interest of the reference marker and the known physical dimension of the reference marker. Finally, the image recognition application determines a stock keeping unit identifier of the identified item in the image based on the pixel-to-physical dimension ratio and a dimension of the region of interest of the identified item.