Image Matching Using Phase Component Extraction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing image matching techniques require significant computational resources due to the need for inverse Fourier transforms, making high-speed image matching challenging.

Innovation Solution

The proposed image matching device employs frequency characteristic acquisition, synthesis of normalized cross power spectra, and complex sine wave determination to assess similarity between images without requiring inverse Fourier transforms, using partial differential calculations and phase angle analysis to derive matching scores efficiently.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If inverse Fourier transform is executed to match two images, then image matching accuracy is improved, but computational time and complexity increase significantly

Engineering Contradiction:
Improveimage matching accuracyVSAvoidcomputational time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent extracts only the phase component from the cross power spectrum by eliminating power components using a weight filter. This extraction of the essential phase information while discarding redundant amplitude information enables accurate image matching without requiring full inverse Fourier transform, thus reducing computational time while maintaining matching accuracy.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

Instead of performing the complete inverse Fourier transform on the entire cross power spectrum, the patent applies partial action by only processing the phase component after filtering. This partial processing approach achieves sufficient matching accuracy for practical applications while dramatically reducing the computational burden of the full transform operation.

Inventive Principle:
Principle #16Partial or excessive action

2Measurement precision

If normalized cross power spectrum is calculated and inverse Fourier transform is executed, then image similarity determination is improved, but device complexity and computational resources increase

Engineering Contradiction:
Improveimage similarity determinationVSAvoidcomputational complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent extracts and utilizes only the phase component of the normalized cross power spectrum for image matching determination. By filtering out power components and focusing solely on phase information, the system achieves accurate similarity determination with reduced computational complexity compared to processing the complete frequency spectrum.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent changes the parameter being analyzed from the complete cross power spectrum to specifically the phase component. This parameter change allows the system to maintain image similarity determination accuracy while significantly reducing the computational resources and device complexity required for processing.

Inventive Principle:
Principle #35Parameter changes

3Measurement precision

If complete frequency characteristic synthesis and inverse Fourier transform are performed, then matching accuracy is improved, but processing speed decreases

Engineering Contradiction:
Improvematching accuracyVSAvoidprocessing speed
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The patent extracts the essential phase information from the synthesized frequency characteristics and eliminates redundant power components. This extraction enables the system to achieve accurate matching results by processing only the necessary phase data, thereby significantly improving processing speed while maintaining matching accuracy.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent applies partial action by performing inverse Fourier transform only on the filtered phase component rather than the complete frequency characteristic synthesis result. This partial processing approach maintains sufficient matching accuracy for practical applications while dramatically improving processing speed by avoiding computation on redundant data.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP3731178B1Image comparison apparatus
Publication Date: 2024.01.31 NEC CORP
  • EP3731178B1 patent drawingFigure 1
  • EP3731178B1 patent drawingFigure 2
  • EP3731178B1 patent drawingFigure 3

AI summary

An image collating device that collates a first image and a second image includes a frequency characteristic acquiring unit, a frequency characteristic synthesizing unit, and a determining unit. The frequency characteristic acquiring unit acquires a frequency characteristic of the first image and a frequency characteristic of the second image. The frequency characteristic synthesizing unit generates a synthesized frequency characteristic by synthesizing the frequency characteristic of the first image and the frequency characteristic of the second image. The determining unit calculates a score indicating a degree to which the synthesized frequency characteristic is a wave having a single period, and collates the first image and the second image based on the score.