Shelf Image Patch Classification for Product Status Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In retail environments, manual detection of product status issues such as restocking and misplacement on shelves is labor-intensive and error-prone due to the fluid nature of inventory and variable imaging conditions like lighting, making automated detection challenging.
Innovation Solution
A method and apparatus using a mobile automation system equipped with image sensors and depth sensors to capture shelf data, decompose images into patches, generate feature descriptors, classify shelf regions, and create masks to identify back-of-shelf areas and gaps, enabling autonomous detection of product status.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual detection methods are used to identify product status issues, then labor flexibility is maintained, but detection accuracy decreases and labor costs increase
Solution Approach 1:
The patent replaces manual visual inspection with an automated imaging system that captures shelf images using image sensors. The system processes these images through algorithms that decompose them into patches, generate feature descriptors, and classify shelf regions to automatically detect product status issues such as out-of-stock conditions and misplacement, thereby substituting mechanical human labor with an automated optical detection system.
2Productivity
If automated detection systems are implemented, then productivity increases, but detection accuracy decreases due to variable imaging conditions
Solution Approach 1:
The patent applies segmentation by decomposing captured shelf images into multiple smaller patches. Each patch is independently processed to generate feature descriptors and classify shelf regions. This segmentation approach allows the system to handle variable imaging conditions more effectively by analyzing local regions separately, improving overall detection accuracy while maintaining automated high-speed processing.
3Measurement precision
If comprehensive image processing is performed to improve detection accuracy, then measurement precision increases, but processing time increases
Solution Approach 1:
The patent implements partial action by selectively processing only relevant image patches rather than performing exhaustive analysis on entire images. The system identifies and focuses computational resources on patches containing potential product status issues, generating feature descriptors and classifications only where needed. This approach maintains high detection accuracy while significantly reducing overall processing time compared to comprehensive full-image analysis.
Data Source
AI summary
A method of detecting a back of a shelf for supporting objects includes: obtaining an image depicting a shelf having a shelf edge and a support surface extending from the shelf edge to a shelf back; decomposing the image into a plurality of patches; for each patch: generating a feature descriptor; based on the feature descriptor, assigning one of a shelf back classification and a non-shelf back classification to the patch; generating a mask corresponding to the image, the mask containing an indication of the classification assigned to each of the patches; and presenting the mask.