Biometric Subject Identification Using Family Correlation Signals

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing fingerprint and voice recognition systems struggle to identify individuals whose biometric data is not present in the database, and voice recognition is not commonly used for subject identification, particularly for family members.

Innovation Solution

A system that identifies subjects by exploiting similarities in fingerprint and voice features between the subject and their family members, using a database of fingerprints and voice recordings to determine correlations and estimate confidence in identification.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If traditional fingerprint recognition systems are used, then identification accuracy is maintained for subjects in the database, but the system fails to identify subjects whose fingerprints are not in the database

Engineering Contradiction:

Improveability to identify subjects not in databaseVSAvoididentification reliability

Core Design Contradiction:

Adaptability or versatilityVSReliability

Solution Approach 1:

The patent introduces family members as intermediary subjects whose fingerprints are stored in the database. When an unknown subject's fingerprint is not found, the system compares the unknown subject's fingerprint against family members' fingerprints to establish familial relationships, thereby enabling identification of subjects not directly in the database while maintaining reliable identification through genetic similarity correlations

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The system changes the identification parameter from direct fingerprint matching to fingerprint similarity correlation analysis. By calculating correlation coefficients between the unknown subject's fingerprint features and family members' fingerprint features, the system can identify subjects based on familial relationships rather than direct database matches, improving adaptability while maintaining reliability through statistical correlation thresholds

Inventive Principle:

Principle #35Parameter changes

2Adaptability or versatility

If voice recognition is used for subject identification, then identification capability is expanded, but the system cannot reliably identify family members using voice similarities

Engineering Contradiction:

Improvevoice-based identification capabilityVSAvoidvoice similarity measurement accuracy

Core Design Contradiction:

Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The patent applies parameter changes by transforming voice recognition from direct identity matching to familial relationship detection. The system extracts voice characteristics (formants, spectral features) and calculates correlation coefficients between the unknown subject's voice and family members' voices, enabling precise measurement of voice similarities to identify family relationships rather than direct matches

Inventive Principle:

Principle #35Parameter changes

3Productivity

If fingerprint features are compared between unknown subject and database subjects, then identification is performed, but the system cannot exploit familial relationship similarities

Engineering Contradiction:

Improveidentification efficiencyVSAvoidfamilial relationship information

Core Design Contradiction:

ProductivityVSLoss of information

Solution Approach 1:

The system performs preliminary action by pre-storing familial relationship information and pre-calculating correlation thresholds in the database. When an unknown subject is presented, the system can immediately compare fingerprint features against family members and utilize pre-established familial relationship patterns, improving identification efficiency while preserving and utilizing familial relationship information rather than losing it

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent implements feedback mechanisms where the system continuously refines identification accuracy by comparing correlation coefficients against predetermined thresholds and adjusting identification confidence levels. The system provides feedback on the strength of familial relationships detected and can request additional verification when correlation strengths are marginal, thereby maintaining high productivity while preventing loss of important familial relationship information

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS12518558B2Subject identification based on correlated self and/or family relation characteristics

Publication Date: 2026.01.06 KINSAME TECHNOLOGIES LTD

AI summary

Identifying subjects based on correlated self and/or family relation (e.g., kinship) characteristics (e.g., fingerprints and/or voice) is described. A previously unseen image or scan or other representation(s) of a fingerprint and/or voice recording of an unknown subject may be used to identify the subject. The previously unseen images of the fingerprint and/or the voice recording may be compared to a database of images of other fingerprints and/or voice recordings. The database of images of other fingerprints and/or voice recordings may include images of other fingerprints and/or other voice recordings of the subject, and/or images of fingerprints and/or voice recordings of family members of the subject. The previously unseen images of the fingerprint and/or the voice recording are correlated with images of other fingerprints and/or voice recordings from the subject, and/or images of fingerprints and/or voice recordings of family members of the subject to identify the subject.