Shelf Edge Content Detection for Accurate Product Label Counting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current retail shelf monitoring systems using cameras and image processing struggle to distinguish between product labels and promotional content on shelf edges, leading to inaccurate counting of products.
Innovation Solution
A method and apparatus that process retail store shelf images to identify and demarcate shelf edge content using geometric patterns and statistical models, allowing product counting systems to differentiate between product labels and shelf edge promotional labels, thereby preventing false counting.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If cameras and image processing technology are used to monitor shelves, then the ability to track product items is improved, but the accuracy of product counting deteriorates due to confusion with promotional material
Solution Approach 1:
The patent segments the shelf image into distinct regions: product label regions and shelf edge promotional material regions. By dividing the image into these separate zones and applying different detection criteria to each, the system can accurately count products while ignoring promotional content that resembles product labels.
Solution Approach 2:
The patent applies different detection algorithms and criteria to different local regions of the shelf image. Product label detection uses one set of parameters while shelf edge promotional material detection uses another, allowing each region to be processed optimally for its specific characteristics and preventing misidentification.
2Adaptability or versatility
If promotional material on shelf edges resembles product labels, then marketing effectiveness is improved, but system reliability deteriorates due to false counting
Solution Approach 1:
The patent exploits asymmetrical differences between product labels and shelf edge promotional material. By detecting geometric patterns, orientations, and spatial relationships that differ between these two types of content, the system can reliably distinguish them even when their visual appearances are similar, maintaining counting accuracy while allowing creative promotional designs.
Data Source
AI summary
An example disclosed method includes receiving an image containing shelf edge content displayed on a shelf edge; determining a location of the shelf edge content and an apparent orientation of the shelf edge content; determining, based on the location of the shelf edge content and the apparent orientation of the shelf edge content, a region in the image corresponding to the shelf edge; and providing an output based on the image that identifies the region in the image corresponding to the shelf edge.