Shelf Image Difference Transmission for Retail State Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The increased communication and processing load on server devices when transmitting captured images to manage product states in retail stores leads to higher power consumption.
Innovation Solution
The system includes an image acquisition unit that acquires a first shelf image and a second shelf image that includes an image acquisition unit that includes an image acquisition unit that acquires a first shelf image and a second shelf image, a region specification unit that specifies a difference region between the first and second shelf images, a cut-out processing unit that cuts out a partial image from the second shelf image, and a transmission processing unit that transmits the first shelf image and the partial image to a reception device.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If captured images are transmitted to the server device for processing, then product state recognition function is provided, but communication amount and server processing load increase
Solution Approach 1:
The patent divides the shelf image into multiple regions and transmits only the difference regions between consecutive images to the server device, rather than transmitting the complete image. This segmentation approach reduces the quantity of data transmitted while maintaining the ability to detect product state changes.
Solution Approach 2:
The patent extracts only the essential information needed for product state recognition by identifying and transmitting only the difference regions between images. This extraction method removes unnecessary data while preserving the critical information about product status changes.
2Reliability
If captured images are transmitted to the server device for processing, then product state recognition function is provided, but server device power consumption increases
Solution Approach 1:
The patent segments the image data transmission by sending only the difference regions rather than complete images, which directly reduces the processing load on the server device and consequently lowers its power consumption while maintaining product state recognition functionality.
Solution Approach 2:
The patent extracts and transmits only the necessary difference region data to the server, reducing the amount of processing required on the server device and thereby reducing its power consumption while preserving the ability to detect product state changes.
3Quantity of substance
If difference region only is transmitted to reduce data amount, then communication load is reduced, but image recognition processing accuracy may be affected
Solution Approach 1:
The patent applies local quality by focusing the transmitted data on specific difference regions where changes occur, rather than transmitting the entire image. This localized approach maintains recognition accuracy for the critical areas while significantly reducing overall data quantity.
Solution Approach 2:
The patent performs preliminary processing at the transmission device to identify and isolate difference regions before transmission. This preliminary action ensures that the transmitted data contains all necessary information for accurate product state recognition while minimizing data volume.
Data Source
AI summary
A transmission device according to the present technology includes an image acquisition unit that acquires a first shelf image captured at a first time for a product shelf on which a product is displayed, and a second shelf image captured for the product shelf at a second time after the first time, a region specification unit that specifies a difference region between the first shelf image and the second shelf image, a cut-out processing unit that cuts out a partial image from the second shelf image so that the difference region is included, and a transmission processing unit that transmits the first shelf image and the partial image to a reception device.