Flow Rack Inventory Counting Using Shelf Image Geometry
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The challenge of efficiently tracking the inventory of multiple flow racks in an industrial production environment, where dozens or hundreds of flow racks are present, due to the sheer number and the need for precise inventory management.
Innovation Solution
A computer-implemented method using image recognition to determine the inventory of a flow rack by analyzing the vertical edge position of the uppermost object on a shelf, leveraging existing image recognition techniques and trigonometric calculations to derive the number of objects based on the shelf's inclination and object dimensions, without requiring physical connections to the rack.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If sensors are equipped on flow racks to track inventory, then measurement precision is improved, but device complexity and cost increase significantly due to the large number of flow racks
Solution Approach 1:
The patent extracts the measurement function from individual flow racks and relocates it to a centralized image processing system. Instead of equipping each rack with sensors, a single camera captures images of multiple racks, and central processing extracts inventory information through image analysis, thereby reducing overall system complexity while maintaining measurement precision
Solution Approach 2:
The patent introduces an intermediary image processing system between the flow racks and the inventory management system. The camera and image processing algorithms serve as mediators that convert visual information into inventory data, eliminating the need for direct sensor integration on each rack and simplifying the overall architecture
2Reliability
If multiple flow racks are monitored with individual sensors, then inventory tracking reliability is improved, but loss of time and cost increase due to installation and maintenance of numerous sensors
Solution Approach 1:
The patent merges the monitoring function across multiple flow racks into a single integrated system. One camera simultaneously captures images of multiple racks, and the processing system analyzes all racks in parallel, thereby maintaining reliable inventory tracking while dramatically reducing installation and maintenance time compared to individual sensor deployment
3Measurement precision
If traditional sensor-based inventory systems are deployed, then measurement precision is improved, but ease of manufacture and deployment deteriorates due to wiring and hardware requirements
Solution Approach 1:
The patent replaces the mechanical sensor-based system with an optical imaging system. Instead of physical sensors requiring wiring and electrical connections, a camera captures visual information that is processed through image analysis algorithms, eliminating complex hardware installation while achieving comparable or superior measurement precision through digital image processing
Data Source
Figure 1a~1c
Figure 2a~2b
Figure 3a~3b
AI summary
A computer-implemented method is for determining an inventory of a flow rack is presented. The flow rack comprises a shelf. The shelf, which may also be referred to as a row of bay compartments spanning the length of the flow rack, is arranged at a tilting angle with respect to a horizontal reference plane. A number of objects is arranged on the shelf. Each of the objects has the same object height and the same object length. The method comprises: Determining, from an image of the shelf, a vertical edge position of an upper edge and/or of a lower edge of an uppermost object of said number of objects arranged on the shelf; and deriving said number of objects arranged on the shelf based on the vertical edge position determined from the image.