Barcodeless Commodity Recognition via Image Matching and Touchscreen Confirmation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional commodity recognition systems rely heavily on barcodes, which are not universally applied, particularly for items like vegetables and fruits, leading to inefficiencies in recognition and data entry during checkout processes.
Innovation Solution
A commodity recognition apparatus utilizing a camera and image processing technology to extract feature data from images of commodities, comparing them to stored reference data to determine recognition codes, enabling recognition and data entry without barcodes through a touchscreen interface.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If barcode scanning is used for commodity recognition, then recognition speed and accuracy are improved, but applicability is limited to commodities with barcodes only
Solution Approach 1:
The system integrates both barcode recognition and image-based recognition capabilities into a single commodity recognition apparatus. The control unit can switch between barcode scanning mode and image processing mode depending on whether the commodity has a barcode, making the system universally applicable to all types of commodities including those without barcodes such as vegetables, fruits, and deli food
2Adaptability or versatility
If image-based recognition is used for commodities without barcodes, then applicability is improved, but recognition accuracy and speed deteriorate
Solution Approach 1:
The system introduces a touchscreen display as an intermediary between the image-based recognition and the final commodity identification. The display shows candidate commodity images and names to the operator, who can visually confirm the correct commodity by comparing it with the captured image, thereby improving recognition accuracy while maintaining applicability to barcodeless commodities
3Adaptability or versatility
If manual selection through touchscreen is used for barcodeless commodities, then applicability is improved, but operation complexity and time consumption increase
Solution Approach 1:
The system performs automatic image capture and automatic candidate commodity search based on the captured image, reducing the operator's workload. The operator only needs to visually confirm the displayed candidate and make a simple selection, rather than manually searching through commodity databases or entering commodity information
4Ease of operation
If candidate commodity display is provided for manual selection, then ease of operation is improved, but recognition speed deteriorates due to additional selection step
Solution Approach 1:
The system displays only the top candidate commodity (or a limited number of candidates) based on image matching results, rather than showing all possible commodities. This partial display approach maintains ease of operation by presenting relevant options to the operator while minimizing the time required for visual confirmation and selection
Data Source
AI summary
A commodity recognition apparatus comprises an image interface, a memory and a processor. The image interface is configured to acquire a commodity image captured by a camera. The memory is configured to store a candidate of a commodity recognized from the commodity image acquired by the image interface. The processor is configured to try to read a commodity recognition code from the commodity image acquired by the image interface and reset the candidate of the commodity stored in the memory if the commodity recognition code is read.