Automatic Feature Registration for Commodity Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing information processing systems for commodity recognition in changing environments, such as supermarkets, face inefficiencies in registering new objects due to the need for manual updates in pre-defined dictionaries, making it time-consuming to adapt to new objects or exclusion objects.
Innovation Solution
An information processing apparatus equipped with an image pickup device, storage device, and processor that allows for the automatic extraction and registration of feature amounts from images, enabling easy addition of new commodities or exclusion objects to dictionaries through a user-friendly interface, facilitating dynamic updates in recognition targets and exclusion objects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If manual registration of new objects in the dictionary is performed, then the recognition system can be updated with new commodities, but the process becomes time-consuming and reduces productivity
Solution Approach 1:
The system automatically extracts feature amounts from images of new commodities and registers them in the dictionary without requiring manual intervention. The image pickup device captures images, the extraction unit automatically extracts feature amounts, and the registration unit stores these features in the dictionary, enabling the system to update itself autonomously
Solution Approach 2:
The manual mechanical process of dictionary registration is replaced by an automated information processing system. The extraction unit uses image processing technology to automatically extract feature amounts from captured images, substituting the manual mechanical registration process with an automated computational approach
2Measurement precision
If the dictionary is updated frequently to include new objects, then the recognition accuracy improves, but the complexity of system operation increases
Solution Approach 1:
The system performs self-updating by automatically extracting feature amounts from images and registering them in the dictionary. This eliminates the need for operators to manually update the dictionary, making the system easy to operate while maintaining high recognition accuracy through frequent updates
Solution Approach 2:
The system performs preliminary extraction of feature amounts from images before registration is needed. By having the extraction unit ready to extract features and the registration unit ready to store them, the system can quickly adapt to new commodities without complex operational procedures
Data Source
AI summary
In accordance with an embodiment, an information processing apparatus includes an image pickup device, a storage device, an operation device, and a processor. The operation device receives selection as to whether to register a feature amount of a reference commodity in a first dictionary file or register a feature amount of an exclusion object in a second dictionary file. The processor extracts a feature amount of an object included in a picked-up image output by the image pickup device. Further, the processor registers the extracted feature amount in the first dictionary file or the second dictionary file in accordance with the selection received by the operation device.