Affine Transformation Parameter Estimation for Biometric Image Alignment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image merging methods require high accuracy in image alignment, leading to user unfriendliness and inaccurate user identification due to improper alignment of biometric features.
Innovation Solution
A computer-implemented method determines an affine transformation to align features of two images by calculating parameters that ensure the transformation exists and is accurate, using a neural network to process images and reduce errors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If image merging methods are applied to combine separate biometric images, then user identification reliability is improved, but alignment accuracy requirement becomes too strict making it user-unfriendly
Solution Approach 1:
The patent changes the parameter space from rigid pixel-perfect alignment to affine transformation parameters. By representing alignment as a set of transform parameters (scaling, rotation, translation) rather than requiring precise pixel correspondence, the system achieves both high identification reliability and user-friendly operation. The affine transformation framework allows flexible parameter adjustments that accommodate natural variations in image capture.
Solution Approach 2:
The patent replaces the mechanical alignment process (manual or algorithmic pixel-level adjustment) with a mathematical model-based approach using affine transformations. This substitution eliminates the need for users to manually align images while maintaining high accuracy through computational geometry and statistical methods for parameter estimation.
2Measurement precision
If high accuracy alignment is required for image merging, then identification accuracy is improved, but errors occur due to lack of proper alignment
Solution Approach 1:
The patent implements feedback mechanisms through iterative optimization processes. The system continuously adjusts transformation parameters based on feature correspondence quality metrics, using feedback from alignment accuracy assessments to refine the transformation model. This ensures that only transformations achieving sufficient precision are accepted for identification.
Solution Approach 2:
The patent performs preliminary actions by pre-processing images to enhance feature detectability and pre-establishing affine transformation models. By preparing the images and transformation framework in advance, the system ensures that subsequent alignment operations achieve high precision without requiring complex real-time adjustments, thereby preventing identification errors.
3Productivity
If affine transformation parameters are determined without verification, then processing speed is improved, but transformation accuracy deteriorates
Solution Approach 1:
The patent applies partial verification by checking only the most critical transformation parameters and using sampling techniques. Instead of exhaustively verifying all possible parameter combinations, the system performs targeted validation on key features and uses statistical confidence intervals to determine sufficiency. This partial action approach maintains high accuracy while significantly improving processing speed.
Solution Approach 2:
The patent changes the verification approach by transforming the validation problem into a statistical parameter estimation problem. Rather than checking each parameter individually, the system uses probabilistic models to assess overall transformation accuracy, adjusting confidence thresholds dynamically. This parameter transformation enables rapid verification without sacrificing precision.
Data Source
AI summary
A computer-implemented method for determining an affine transformation for transforming a second image so that features of the second image coincide with corresponding features of a first image, the computer-implemented method comprising obtaining the second image comprising the features and obtaining the first image comprising the corresponding features, processing the images by a transformation calculator, thereby determining an indication of whether the affine transformation exists and determining parameters defining the affine transformation, and outputting the affine transformation, comprising outputting the indication of whether the affine transformation exists and outputting a data structure indicative of the parameters defining the affine transformation.