Scoring Imaging Methods and Trained Models for Data Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data classification techniques face challenges in accurately classifying data using learned classification models, as they require manual or automatic examination of suitable imaging methods and trained models, leading to increased operation loads and time due to the numerous methods and models available.
Innovation Solution
An information processing method that calculates a score for each imaging method and trained model based on the distribution of feature values in a feature value space, allowing for the identification of the most suitable combination for accurate data classification with reduced operational effort.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual or automatic examination of suitable imaging methods and trained models is performed, then accurate data classification can be achieved, but operation loads and time increase due to the numerous methods and models available
Solution Approach 1:
The patent pre-calculates scores for multiple imaging methods and trained models before actual classification tasks. These scores represent pre-evaluated suitability metrics that allow rapid selection during operation, eliminating the need for time-consuming examinations when new classification tasks arise.
Solution Approach 2:
The patent transforms the evaluation of imaging methods and trained models into quantitative score parameters. By converting qualitative assessments into measurable score values, the system enables efficient comparison and selection among multiple methods and models without manual examination.
2Measurement precision
If manual or automatic examination of suitable imaging methods and trained models is performed, then accurate data classification can be achieved, but operation loads increase due to the numerous methods and models available
Solution Approach 1:
The system performs preliminary evaluation and scoring of imaging methods and trained models in advance, storing these results for quick retrieval. This eliminates the need for operators to manually examine multiple methods and models during actual classification operations, significantly reducing operation loads.
Solution Approach 2:
The system automatically calculates scores and determines the most suitable imaging method and trained model combinations without requiring manual intervention. This self-evaluating mechanism reduces operation loads by eliminating the need for operators to examine and select among numerous methods and models.
3Measurement precision
If multiple imaging methods and trained models are evaluated to find the optimal combination, then classification accuracy improves, but device complexity increases
Solution Approach 1:
The patent simplifies the complexity of evaluating multiple imaging methods and trained models by transforming the evaluation into a scoring parameter system. Each method and model receives a quantitative score that reflects its suitability, allowing the system to manage complexity through standardized parameter representation rather than complex evaluation procedures.
Data Source
AI summary
An information processing method implemented by a computer, the information processing method includes: acquiring an image group generated by imaging a data group according to each of a plurality of imaging methods; for each of the acquired image groups, calculating a score of the imaging method used to generate the image group, based on distribution of a first feature value group in a feature value space, and distribution of a second feature value group in the feature value space, the first feature value group being a plurality of feature values output when the image group is input to a trained model outputting feature values corresponding to input images, the second feature value group being a plurality of feature values output when a reference image group is input to the trained model; and outputting the score of the imaging method, the score being calculated for each of the image groups.