Partition Pattern Extraction for Object Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing pattern recognition systems face high processing loads when identifying multiple objects, as they require calculating and comparing feature values for each object, making it burdensome and computationally intensive.
Innovation Solution
An information processing device that extracts partition patterns from captured images, comparing them to prestored patterns to identify objects, reducing the need for extensive image analysis by using a captured image receiver, partition pattern extractor, and object recognizer, and optionally includes a selection operation recognizer, request sender, and information output controller to retrieve information related to selected areas.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If feature values are calculated and compared for each object in traditional pattern recognition systems, then object identification accuracy is improved, but processing load increases significantly
Solution Approach 1:
The patent segments the object identification process into two stages: first extracting partition patterns (structural framework) from the captured image, then comparing only these patterns against prestored templates. This segmentation allows the system to focus computational resources on comparing simplified structural representations rather than analyzing complete feature sets for multiple objects, thereby reducing processing load while maintaining identification accuracy.
Solution Approach 2:
The patent extracts only the essential partition pattern information from captured images - the structural framework that defines object boundaries and internal divisions. By taking out and isolating this critical structural information for comparison, the system eliminates the need to process and compare comprehensive feature values of all potential objects, significantly reducing computational complexity while preserving identification capability.
2Adaptability or versatility
If feature values of all objects are obtained in advance for pattern recognition, then comprehensive object identification capability is improved, but data acquisition burden increases
Solution Approach 1:
The patent uses prestored partition pattern templates as simplified copies or representations of objects, rather than storing complete feature sets. These templates capture the essential structural patterns that define objects, allowing the system to maintain comprehensive object identification capability while dramatically reducing the data acquisition and storage burden associated with traditional feature value databases.
Data Source
AI summary
For example, a captured image including an image to be processed is received. In the image to be processed, various items are separately displayed in a plurality of areas. Subsequently, a partition pattern that partitions the image to be processed, which is included in the received captured image, into a plurality of areas is extracted. An object to be processed is recognized by comparing the extracted partition pattern with one or more partition patterns prestored in a storage unit and then identifying a partition pattern corresponding to the partition pattern extracted from the image to be processed. This makes it possible to recognize an object to be processed under a relatively light processing load.