Computer Vision Shelf Auditing for Real-Time Restocking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for restocking store shelves are inefficient and prone to errors, leading to delayed restocking and lost sales due to the lack of real-time inventory updates and manual oversight, which can result in customer frustration and loss of loyalty.
Innovation Solution
A computer vision system using a machine-learning model trained on images of store shelves to identify missing items and calculate the quantity needed for restocking, providing real-time reports to store personnel for timely restocking.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If manual checking by employees is used, then the system is simple to implement, but it is inefficient and time-consuming leading to delayed restocking
Solution Approach 1:
The patent replaces the manual mechanical checking process with an automated computer vision system using cameras and machine learning algorithms. The system captures images of shelves, processes them through trained models to detect empty spaces and identify products, and automatically generates restocking reports, eliminating the need for manual visual inspection while significantly improving restocking speed and accuracy.
Solution Approach 2:
The system enables self-monitoring and self-reporting of inventory status. The computer vision system automatically detects when products are out of stock and generates restocking alerts without human intervention, allowing the store to monitor its own inventory state and trigger restocking actions autonomously based on detected conditions.
2Measurement precision
If transactional data from POS systems is used, then the system is simple to implement, but the data does not update in real time and is inconsistent with actual shelf inventory
Solution Approach 1:
The patent implements continuous monitoring of shelf inventory through constantly active cameras that capture images at regular intervals. This continuous action ensures that the system always has current visual data of shelf stock levels, eliminating the delays and inconsistencies associated with periodic manual checks or non-real-time POS data updates.
Solution Approach 2:
The system establishes a feedback loop where the computer vision system continuously monitors actual shelf inventory, compares it with expected stock levels, and provides real-time feedback when discrepancies are detected. This feedback mechanism enables immediate detection of out-of-stock conditions and triggers timely restocking actions, ensuring inventory accuracy matches actual shelf conditions.
3Reliability
If manual restocking is used, then the process is simple, but it results in popular items missing from shelves for longer than necessary
Solution Approach 1:
The system performs preliminary detection and reporting of empty spaces and out-of-stock conditions before they become critical. By continuously monitoring shelf inventory and generating restocking alerts as soon as products are detected to be missing, the system enables restocking actions to be taken in advance, preventing popular items from remaining out of stock for extended periods and ensuring continuous product availability.
Data Source
AI summary
A machine-learning model (MLM) is trained to identify a given item identifier for an item and shelf dimensions of an empty space associated with the item from training images of a shelf. After training, real-time images of the shelf are provided as input to the MLM and the output provided by the MLM includes empty space identifiers, dimensions or pixel coordinates for each empty space identifier, and an item identifier for each empty space identifier. A quantity of each item identifier is determined based on known shelf dimensions that the corresponding item should occupy on a fully stocked shelf and based on the corresponding empty space dimensions for the empty space associated with the item. A real-time report is sent to store personnel and/or published on a website monitored by the store personnel. The report identifies the items, the shelves, and restocking item quantities that need restocked in the store.