Label Issuing Device Using Image Recognition for Product Identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In supermarkets and food processing plants, part-time staff without professional knowledge often struggle to identify products accurately, leading to confusion when labeling items like fish fillets, as they are unsure of the type of fish, which can result in incorrect labeling and potential sales of defective or improperly prepared products.
Innovation Solution
A label issuing device equipped with imaging means, object recognition technology, and a PLU file system that captures product images, recognizes specific objects, reads information from a database, and prints labels with accurate product details, including name and price, using a dual printer setup with a display and weighing scale for efficient operation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If manual labeling by part-time staff is used, then labor cost is reduced, but labeling accuracy deteriorates due to lack of professional knowledge
Solution Approach 1:
The system enables self-service labeling by having the imaging device and recognition system automatically identify products and generate labels without requiring staff to manually identify or research product information. The labeling process serves itself through automated image capture, object recognition, and database querying.
Solution Approach 2:
The manual mechanical process of identifying products and researching their information is replaced by an automated system combining imaging devices, object recognition algorithms, and database queries. This substitution eliminates the need for staff to have professional knowledge while maintaining high labeling accuracy.
2Measurement precision
If automated imaging and recognition systems are introduced, then labeling accuracy is improved, but device complexity increases
Solution Approach 1:
The imaging device serves multiple functions: capturing product images, enabling object recognition, and providing visual reference for the labeling process. This multi-functionality reduces the need for separate specialized devices, thereby limiting the increase in overall system complexity while maintaining high labeling accuracy.
3Measurement precision
If image comparison with database is performed, then product identification accuracy is improved, but processing time increases
Solution Approach 1:
Product images and their corresponding information are pre-loaded into the database before the labeling process. This preliminary action allows the system to quickly compare captured images against pre-indexed data rather than performing real-time analysis, significantly reducing processing time while maintaining high identification accuracy.
Data Source
AI summary
A label issuing device includes an imaging portion, an image output portion that outputs images captured by the imaging portion, an object recognition portion that recognizes a specific object by reading a feature amount of the output images, an information reading portion that reads information for the recognized object from a file storing information related to the object, and a label issuing portion that prints the read information for the object on a label and issues the label.