Dynamic Image Search Using Principal Component Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image search systems face challenges in accurately determining the search area based on image data, making it difficult to efficiently find desired images that match user preferences.
Innovation Solution
A searching system that uses principal component analysis to generate and output new item candidates based on the positional relationship and component value distribution between a reference image and selected images, allowing for dynamic adjustment of the search range according to user preferences.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the search area is determined based on image data of the reference image, then the search process can be initiated, but the search accuracy is insufficient to efficiently find desired images
Solution Approach 1:
The system uses feedback from user selections to iteratively refine the search area. After displaying candidate images and receiving user feedback, the system calculates the positional relationship between the reference image and selected images, then updates the component value distribution to define a new, more accurate search area that better reflects user preferences.
Solution Approach 2:
The search area is made dynamic rather than static. The component value distribution and search boundaries are automatically adjusted based on user feedback and the calculated positional relationships between images. This dynamic adaptation allows the search area to expand or contract in different principal component directions based on actual user preferences.
2Loss of information
If various sample images are provided to the user for selection, then user preference information can be acquired, but the search area determination remains inaccurate
Solution Approach 1:
User selections of sample images provide feedback that is systematically processed. The control unit calculates the positional relationship between the reference image and selected images in the principal component space, using this feedback to refine the component value distribution and improve search area accuracy in subsequent iterations.
Solution Approach 2:
The system changes parameters of the search area based on user feedback. By calculating positional relationships and updating component value distributions, the search parameters (boundaries, center, spread) are dynamically adjusted to better match user preferences while maintaining accurate search area determination.
Data Source
AI summary
Provided are a searching system, a searching method, and a searching program for searching for an item desired by a user. An assistance server comprises a control unit—that is connected to a user terminal. The control unit outputs, to the user terminal, a first item including a first component value in a plurality of principal components composing items, and a plurality of item candidates including component values that are different from the first component value. The control unit specifies a second item selected in the user terminal from among the plurality of item candidates. The control unit calculates, for each of principal component values, the positional relationship between the first component value and a second component value of the second item. The control unit calculates, for each of the principal component values, a distribution of the component values according to the positional relationship. The control unit newly generates, for the second item, a plurality of item candidates on the basis of the distribution of the component values. The new item candidates are output to the user terminal.