Object Authentication via Blueprint-Fingerprint Statistical Dependency

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing authentication methods for physical objects face challenges such as high intra-class variability, lack of accurate mathematical models, and vulnerability to advanced attacking strategies, including adversarial examples and copy attacks.

Innovation Solution

A method for authenticating physical objects based on a statistical dependence between blueprints and fingerprints, using a parametrized and learnable approach to establish a joint distribution between blueprint-fingerprint pairs, allowing for accurate prediction and simulation of fingerprints from blueprints without requiring enrollment of fingerprints from each physical object.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If fingerprint enrollment is performed for each physical object to improve authentication reliability, then authentication accuracy is improved, but cost and time consumption increase

Engineering Contradiction:

Improveauthentication accuracyVSAvoidenrollment time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The patent performs preliminary action by training the authentication model in advance using a dataset of blueprint-fingerprint pairs before actual authentication. This pre-training phase captures the statistical dependencies between blueprints and fingerprints, enabling the system to authenticate objects without requiring individual fingerprint enrollment for each object, thus resolving the contradiction between authentication reliability and enrollment time

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent uses blueprint images as substitutes (copies) for actual fingerprint enrollment. Instead of requiring the original fingerprint data from each physical object, the system generates or retrieves corresponding blueprint images that contain sufficient statistical information for authentication, eliminating the need for time-consuming fingerprint capture while maintaining authentication accuracy

Inventive Principle:

Principle #26Copying

2Reliability

If fingerprint enrollment is performed for each physical object to improve authentication reliability, then authentication accuracy is improved, but cost increases

Engineering Contradiction:

Improveauthentication accuracyVSAvoidauthentication cost

Core Design Contradiction:

ReliabilityVSLoss of energy

Solution Approach 1:

The system performs preliminary model training using a comprehensive dataset of blueprint-fingerprint pairs, capturing statistical dependencies in advance. This pre-computed model can then authenticate objects without requiring expensive individual fingerprint enrollment infrastructure, reducing operational costs while maintaining high authentication accuracy through the learned statistical relationships

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent replaces expensive fingerprint enrollment and storage infrastructure with cheaper blueprint image processing. By using blueprint images that can be obtained or generated more economically than actual fingerprint captures, the system maintains authentication reliability while significantly reducing the cost associated with fingerprint enrollment and database management

Inventive Principle:

Principle #26Copying

3Ease of operation

If traditional authentication methods are used to simplify the process, then ease of operation is improved, but vulnerability to attacks increases

Engineering Contradiction:

Improveauthentication simplicityVSAvoidattack vulnerability

Core Design Contradiction:

Ease of operationVSObject-affected harmful factors

Solution Approach 1:

The patent transforms the authentication approach by changing from direct fingerprint matching to statistical dependency analysis between blueprints and fingerprints. This parameter change involves using learned statistical models and probability distributions rather than simple comparison algorithms, maintaining operational simplicity while providing robustness against attacks through the complexity of the statistical relationships that are difficult to exploit

Inventive Principle:

Principle #35Parameter changes

4Measurement precision

If detailed fingerprint analysis is performed to improve measurement precision, then authentication accuracy is improved, but device complexity increases

Engineering Contradiction:

Improvefingerprint analysis precisionVSAvoidsystem complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent introduces statistical models and probability distributions as intermediaries between raw fingerprint data and authentication decisions. Instead of directly analyzing complex fingerprint details, the system uses learned statistical relationships from blueprint-fingerprint pairs as a mediator, simplifying the analysis process while maintaining measurement precision through the information captured in the statistical dependencies

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS12293007B2Object authentication using digital blueprints and physical fingerprints

Publication Date: 2025.05.06 UNIVERSITY OF GENEVA

AI summary

A method of object authentication based on digital blueprints and physical fingerprints comprising the steps of acquiring a set of training blueprints and fingerprints, training, object enrollment and object authentication. The method uses a pair of a mapper realized as an encoder and a decoder and a set of multi-metric scores originating from the decomposition of mutual information and applied to both the output of the encoder and decoder and producing a feature vector for a one-class classifier. The method is trained only on the original physical objects and does not use any fakes for reliable authentication.