Image Registration Accuracy via Displacement Field Matching Degrees
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In image diagnosis, the accuracy of corresponding point matching between images captured in different modalities, at various times, and from different body positions is uncertain due to user input errors, affecting the correctness of image registration and analysis.
Innovation Solution
An information processing apparatus that calculates and displays the matching degrees of corresponding point candidates by generating displacement fields under different regularization conditions, allowing users to visually recognize and correct errors, thereby improving the accuracy of image registration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If user manually specifies corresponding points between images, then positioning accuracy can be improved, but the correctness of corresponding points becomes uncertain due to user input errors
Solution Approach 1:
The system calculates matching degrees for multiple candidate corresponding points and provides feedback to the user through display. This allows users to identify and correct incorrect corresponding points based on the calculated matching degrees, thereby improving the reliability of corresponding points while maintaining positioning accuracy
Solution Approach 2:
The system automatically calculates matching degrees and identifies candidate corresponding points without requiring manual verification of each point's correctness. This self-evaluation mechanism reduces user input errors while preserving the ability to achieve high positioning accuracy through automatic correction
2Measurement precision
If multiple displacement fields are calculated under different regularization conditions, then the accuracy of evaluating corresponding points is improved, but the computational complexity increases
Solution Approach 1:
The system calculates multiple displacement fields under different regularization conditions to obtain a sufficient number of candidate corresponding points with reliable matching degrees. By calculating more displacement fields than the minimum single field, the system ensures accurate evaluation of corresponding points while managing computational complexity through targeted calculations
Data Source
AI summary
An information processing apparatus includes a processor and a memory containing instructions that cause the processor to perform operations. The operations include obtaining a group of candidates of corresponding points which are pairs of feature points in a first space and feature points in a second space, calculating a plurality of displacement fields under a plurality of different regularization conditions based on the candidate group, calculating a feature point group which has been subjected to a plurality of displacements by displacing the feature points in the first space by the plurality of displacement fields, and calculating matching degrees of the corresponding points included in the candidate group in accordance with the feature point group which has been subjected to the plurality of displacements.