Adjacent Item Filtering Using Vertical Retail Compartment Boundaries
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current computer vision (CV) item recognition systems in retail environments suffer from inaccurate item and location mapping due to false positives, making manual verification inefficient and costly in large facilities.
Innovation Solution
A system that utilizes a filter manager to analyze CV item recognition results, identify vertical members of storage compartments, and track them across images to filter out adjacent items, improving item-to-location mapping accuracy by differentiating between target and adjacent compartments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If computer vision item recognition is used to automatically identify products and locations, then productivity is improved, but measurement precision deteriorates due to false positives in item-to-location mapping
Solution Approach 1:
The patent introduces vertical member detection as an intermediary step between image capture and item-to-location mapping. The system detects vertical members (shelving structures) to define compartment boundaries, which serve as a mediator to correctly associate items with their locations. This intermediary detection mechanism resolves the false positive problem by establishing proper spatial relationships before final mapping occurs.
Solution Approach 2:
The patent segments the storage structure into distinct compartments by detecting vertical members that define compartment boundaries. This segmentation allows the system to process and map items to specific locations more accurately by treating each compartment as a separate entity, thereby improving measurement precision while maintaining automated productivity.
2Measurement precision
If manual verification is performed to correct CV errors, then measurement precision is improved, but productivity deteriorates due to time-consuming manual inspection
Solution Approach 1:
The patent implements self-service by enabling the computer vision system to automatically correct its own mapping errors through vertical member detection and compartment definition. The system performs self-verification by using detected vertical members to validate and correct item-to-location mappings without requiring manual intervention, thus maintaining high measurement precision while preserving productivity.
3Measurement precision
If comprehensive item recognition is performed across all storage areas, then measurement precision is improved, but use of energy deteriorates due to processing large volumes of data
Solution Approach 1:
The patent performs preliminary action by detecting vertical members and defining compartment boundaries before conducting full item recognition and mapping. This preliminary structural detection creates a framework that guides subsequent item processing, allowing the system to maintain high measurement precision while reducing computational energy consumption by avoiding unnecessary processing of items outside defined compartments.
Data Source
AI summary
Examples provide a system for filtering the contents of adjacent item storage compartments from item recognition results obtained using computer vision (CV) for more accurate mapping of item locations within a retail environment. A filter manager selects a vertical member and tracks it throughout a series of images generated by an image capture device. The selected vertical member defines at least a portion of a target compartment in each image. The vertical member can be a display case door or a vertical steel bar defining the side of an item storage compartment. A set of adjacent items located on each side of the target compartment is filtered from the item recognition results. The target compartment location is determined based on a location tag of the target compartment. The items remaining after filtering are mapped to the target compartment location while reducing CV item location false positives.