Fixed-Camera Stock Keeping With Planogram-Based Product Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current stock keeping methods in retail stores face challenges in accurately and efficiently tracking inventory using fixed cameras, especially with low object-level resolutions and limited network bandwidth, and require more effective methods to identify product types and quantities across varying lighting conditions and resolutions.
Innovation Solution
A method utilizing a computer system that accesses images from fixed cameras and a robotic system to estimate inventory structure segments, identify product units, and update real-time stock conditions by leveraging a planogram to match features with product models, even at low resolutions, and generate restocking prompts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If fixed cameras with low resolution are used to reduce network bandwidth requirements, then network bandwidth consumption is reduced, but measurement precision of product identification deteriorates
Solution Approach 1:
The system performs preliminary actions by capturing images at multiple resolutions and pre-processing them to extract key features before transmission. This allows the full-resolution images to be processed locally while only essential data is transmitted over the network, reducing bandwidth requirements while maintaining identification accuracy.
Solution Approach 2:
The image processing is segmented into multiple stages: initial low-resolution filtering to identify regions of interest, followed by targeted high-resolution analysis of only those specific areas. This segmentation allows the system to maintain high measurement precision for product identification while transmitting minimal data over the network.
2Measurement precision
If multiple high-resolution images are transmitted for accurate inventory tracking, then measurement precision improves, but network bandwidth requirements increase
Solution Approach 1:
The system extracts only the essential features and information from high-resolution images that are necessary for inventory tracking. By taking out and transmitting only these critical data elements rather than the complete high-resolution images, the system maintains measurement precision while dramatically reducing network bandwidth consumption.
Solution Approach 2:
The system applies partial action by processing and transmitting only the portion of image data that is sufficient for accurate inventory tracking. Rather than transmitting complete high-resolution images, it transmits selectively processed data that provides the necessary measurement precision without the excess bandwidth consumption of full image transmission.
3Device complexity
If feature matching is performed on low-resolution images to reduce processing load, then device complexity is reduced, but measurement precision deteriorates
Solution Approach 1:
The system performs preliminary feature extraction and enhancement on low-resolution images before the main matching process. This preliminary action prepares the data in a way that maintains measurement precision while keeping the overall processing complexity manageable by breaking down the complex task into simpler sequential steps.
Solution Approach 2:
The system changes parameters of the image data through various processing techniques such as contrast enhancement, feature highlighting, and adaptive filtering. These parameter changes improve the effectiveness of feature matching on low-resolution images, maintaining measurement precision while working within the constraints of reduced processing complexity.
Data Source
AI summary
A method for stock keeping in a store includes: accessing an image captured by a fixed camera within the store; retrieving a field of view of the fixed camera; estimating a segment of an inventory structure in the store depicted in the image based on a projection of the field of view onto a planogram of the store; identifying a set of slots within the inventory structure segment; retrieving a product model representing a set of visual characteristics of a product type assigned to a slot, in the set of slots, by the planogram; extracting a constellation of features from the image; if the constellation of features approximates the set of visual characteristics in the product model, detecting presence of a product unit of the product type occupying the inventory structure segment; and representing presence of the product unit, occupying the inventory structure segment, in a realogram.