Image Selection Apparatus Using Query Information Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image selection systems face challenges in efficiently notifying users of search results when multiple images are queried, as they struggle to effectively display and prioritize similar images based on their similarity to the query images.
Innovation Solution
An image selection apparatus and method that acquires query information for multiple query images, selects similar images based on similarity, and controls the display of these images on a unit, setting their position or order using the number of query images that satisfy a reference, allowing users to efficiently identify relevant results.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If multiple query images are used to search for similar images, then the information related to search results increases, but it becomes difficult to efficiently notify users of the search results
Solution Approach 1:
The patent segments the display of search results by organizing similar images into multiple groups based on their similarity to different query images. Each group is displayed separately with clear visual distinction, allowing users to process information in manageable segments rather than overwhelming them with all results simultaneously. This segmentation resolves the contradiction by maintaining comprehensive information while improving notification efficiency through structured presentation.
Solution Approach 2:
The patent introduces a new dimension for organizing search results by displaying similar images in multiple rows or layers, where each row represents a different group of similar images associated with specific query images. This dimensional organization transforms the flat list of results into a multi-level structure, enabling users to efficiently navigate and understand the relationships between query images and their corresponding similar images while preserving all relevant information.
2Loss of information
If similar images are selected based on similarity to at least one query image, then the search results are comprehensive, but the display becomes complex and hard to interpret
Solution Approach 1:
The patent segments similar images into distinct groups based on their similarity relationships with different query images. Each group is visually separated and labeled to indicate which query image(s) it corresponds to. This segmentation maintains the completeness of search results by including all similar images while reducing display complexity through organized grouping and clear visual hierarchy.
Solution Approach 2:
The patent employs multi-row display where each row represents a different similarity group. This dimensional approach allows comprehensive display of all similar images while organizing them in a structured format that reduces perceived complexity. Users can easily interpret the display by understanding that each row represents a specific relationship between query images and similar images.
3Adaptability or versatility
If the number of query images increases, then the search coverage improves, but the user's ability to understand and interact with results deteriorates
Solution Approach 1:
The patent segments the display to show similar images in multiple rows, where each row corresponds to a specific query image or group of query images. This segmentation allows the system to handle an increasing number of query images while maintaining user understanding by presenting results in organized, manageable sections rather than a single overwhelming list.
Solution Approach 2:
The patent uses multi-row display architecture that adds a dimensional organization to search results. As the number of query images increases, the system automatically distributes similar images across multiple rows, maintaining a consistent and interpretable layout. This dimensional approach preserves search coverage while preventing user overload by structuring the information in a visually manageable format.
Data Source
AI summary
A query acquisition unit (610) acquires a plurality of pieces of query information. The query information is information generated for each of a plurality of query images, and indicates a feature of the query image. A similar image selection unit (620) selects, as a similar image by using the plurality of pieces of query information, an image whose degree of similarity to at least one of the query images satisfies a reference. A display control unit (630) displays the similar image selected by the similar image selection unit (620) on a display unit. When the similar image selection unit (620) selects a plurality of the similar images, the display control unit (630) sets a display position or a display order of each of the plurality of similar images on the display unit by using the number of the query images in which the similar image satisfies a reference.