POS Commodity Identification via Image Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Point-of-sale (POS) systems, especially self-service systems, face inefficiencies in commodity code reading due to the need for operators to manually scan barcodes, which is time-consuming and burdensome, particularly for inexperienced users.
Innovation Solution
A POS apparatus and system that includes a control unit connected to an imaging unit, capable of identifying commodities using both barcode symbols and external appearance features from captured images, thereby reducing the reliance on manual barcode scanning and improving operability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If manual barcode scanning is used, then the system can identify commodities, but the operation becomes slow and burdensome
Solution Approach 1:
The patent replaces the mechanical manual scanning process with an automated image recognition system. The imaging unit captures images of commodities, and the control unit automatically identifies barcodes and external appearance features, eliminating the need for operators to manually hold and scan barcodes, thus increasing productivity while reducing operational burden
Solution Approach 2:
The system enables self-service commodity identification where the imaging unit and control unit automatically perform the identification function without requiring operator intervention. The control unit autonomously processes images, identifies commodities through barcode recognition and external feature analysis, and registers them in the POS system, allowing the system to serve itself rather than requiring manual operation
2Measurement precision
If barcode scanning is required, then commodity identification is accurate, but inexperienced users struggle to perform the operation
Solution Approach 1:
The patent introduces an intermediary image processing layer between the camera and commodity identification. The control unit captures images through the imaging unit, processes them to extract both barcode information and external appearance features, and uses these as intermediaries to accurately identify commodities. This intermediary process makes the system more user-friendly while maintaining accuracy
Solution Approach 2:
The control unit performs multiple identification functions: it recognizes barcodes, extracts external appearance features, and identifies commodities. This multi-functional approach ensures that even if barcode scanning fails or is difficult for inexperienced users, the system can still accurately identify commodities through alternative methods, thereby maintaining both accuracy and ease of operation
3Productivity
If the system relies on barcode reading, then commodity registration is straightforward, but the process becomes time-consuming
Solution Approach 1:
The imaging unit continuously captures images of commodities as they are placed on the conveyor belt, and the control unit continuously processes these images to identify commodities. This continuous operation eliminates interruptions and waiting times associated with manual barcode scanning, maintaining productive action throughout the registration process and reducing time loss
Solution Approach 2:
The system performs preliminary image capture and preliminary identification processing as commodities move along the conveyor belt, before the registration is finalized. The control unit extracts barcode information and external features in advance, preparing the identification results ahead of time, which speeds up the overall registration process and reduces time consumption
Data Source
AI summary
According to one embodiment, a point-of-sale apparatus includes a connection interface connectable to a point-of-sale (POS) terminal and a control unit connectable to an imaging unit. The control unit is configured to acquire a captured image of a commodity from the imaging unit, attempt to identify the commodity in the image using a code symbol attached to the commodity, attempt to identify the commodity based on external appearance feature data of the commodity in in the captured image, and output commodity identification information for the commodity to the POS terminal according to either the code symbol or the external feature data.