Learning Model Generation for Image Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image correction technologies require a large number of learning models for diverse imaging conditions, making them complex and inefficient.
Innovation Solution
A learning model generation apparatus that classifies image information using evaluation values derived from imaging conditions such as ISO sensitivity, shutter speed, and iris diaphragm, allowing for the generation of fewer learning models by grouping similar conditions together.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a learning model is generated for each imaging condition, then the accuracy of image correction is improved, but the number of learning models increases and complexity increases
Solution Approach 1:
The patent merges multiple imaging conditions into broader categories by establishing hierarchical relationships between conditions. Learning models trained on specific conditions can be applied to related conditions through the hierarchy, reducing the total number of models needed while maintaining correction accuracy across diverse imaging scenarios.
Solution Approach 2:
The patent creates learning models with universal applicability by designing them to handle multiple imaging conditions through hierarchical relationships. A single learning model can serve multiple functions across different imaging conditions by leveraging the established hierarchy, reducing the need for separate specialized models for each condition.
2Adaptability or versatility
If diverse imaging conditions are handled with separate learning models, then the adaptability to different conditions is improved, but the complexity and resource requirements increase
Solution Approach 1:
The patent segments the diverse imaging conditions into hierarchical groups, organizing them from specific to general categories. This segmentation allows the system to handle diverse conditions adaptably by selecting appropriate models from different hierarchical levels, while reducing overall complexity through structured organization rather than treating each condition independently.
Solution Approach 2:
The patent changes the parameter of model selection from fixed one-to-one mapping to dynamic hierarchical selection. By introducing evaluation values and hierarchical relationships, the system can adapt to different imaging conditions by selecting models at appropriate hierarchical levels, maintaining versatility while reducing the effective number of models needed.
Data Source
AI summary
A learning model generation apparatus includes: a processor configured to obtain captured image data and plural setting values which are set for each imaging condition in a case where the image data is captured and have dependency relationships with one another; calculate an evaluation value for classifying image information which is information obtained from the image data by using the plural setting values; classify the image information based on the evaluation value; and generate a learning model for each classification by using the image information.