Image Matching Device Resolution Regulation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing image matching techniques using phase-only correlation struggle to appropriately regulate the resolution of correlation coefficient maps, leading to inconsistent matching accuracy and failure in distinguishing identical from different individuals due to inappropriate resolution settings based on cross power spectrum frequencies.

Innovation Solution

An image matching device and method that automatically determines the minimum necessary resolution for correlation coefficient maps separately from frequency components, allowing for precise regulation of resolution to enhance matching accuracy and reduce calculation data while ensuring speedy processing.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If the resolution of correlation coefficient map is determined based on frequency components from cross power spectrum, then the calculation amount is reduced, but the matching accuracy becomes inconsistent and unreliable

Engineering Contradiction:
Improvecalculation amountVSAvoidmatching accuracy
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The patent separates the resolution determination from frequency component extraction. The resolution of correlation coefficient map is independently determined based on image characteristics rather than being tied to frequency band selection, allowing independent optimization of both calculation efficiency and matching accuracy

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces a parameter (resolution) that can be independently adjusted from frequency components. By changing the resolution parameter separately from frequency band selection, the system can optimize matching accuracy without being constrained by frequency-based resolution determination

Inventive Principle:
Principle #35Parameter changes

2Reliability

If low-frequency components are extracted to increase robustness against noise, then the resolution of correlation coefficient map decreases, but the matching reliability becomes inconsistent

Engineering Contradiction:
Improverobustness against noiseVSAvoidresolution of correlation coefficient map
Core Design Contradiction:
ReliabilityVSMeasurement precision

Solution Approach 1:

The patent decouples the noise robustness function (achieved through low-frequency extraction) from the resolution function. By determining resolution independently from frequency component selection, the system maintains noise robustness while ensuring adequate resolution for reliable matching

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent enables dynamic adjustment of resolution based on actual matching requirements rather than static frequency-based determination. This allows the resolution to be adapted to maintain reliability while accommodating varying noise conditions and matching demands

Inventive Principle:
Principle #15Dynamics

Data Source

PatentEP3764319B1Image comparison device
Publication Date: 2023.11.22 NEC CORP
  • EP3764319B1 patent drawingFigure 1
  • EP3764319B1 patent drawingFigure 2
  • EP3764319B1 patent drawingFigure 3

AI summary

An image matching device that performs matching between a first image and a second image includes: a frequency characteristic acquisition unit configured to acquire a frequency characteristic of the first image and a frequency characteristic of the second image; a frequency characteristic synthesizing unit configured to synthesize the frequency characteristic of the first image and the frequency characteristic of the second image to generate a synthesized frequency characteristic; a determination unit configured to perform frequency transformation on the synthesized frequency characteristic to calculate a correlation coefficient map whose resolution coincides with a target resolution, and perform matching between the first image and the second image based on a matching score calculated from the correlation coefficient map; and a regulation unit configured to regulate the target resolution based on the matching score.