Font Authenticity Verification Using Embedded Space Representations

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Manual verification of document authenticity is labor-intensive and prone to errors, especially when dealing with rare or new document types, and existing automated systems face challenges in handling various document formats and anti-forgery measures across different countries.

Innovation Solution

A computer-implemented method using embedded space representations to compare characters in a sample image with a predefined font, involving training character extraction, segmentation, and convolutional neural networks to assess font authenticity, enabling accurate identification even with imaging artefacts and adapting to new document types.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If manual verification is used to assess document authenticity, then the assessor can take multiple factors into account, but the process becomes labor-intensive and prone to uncertainty

Engineering Contradiction:

Improveassessment accuracyVSAvoidprocessing efficiency

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

The patent replaces manual mechanical verification with an automated computer-implemented system that uses image processing and font analysis algorithms to assess document authenticity, eliminating human labor while maintaining assessment capability

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system creates digital copies of document images and analyzes them through computational methods, allowing multiple assessments of the same document without additional manual effort and enabling consistent reproduction of verification results

Inventive Principle:

Principle #26Copying

2Productivity

If automated barcode scanning is used to verify documents, then processing efficiency improves, but the system cannot handle all document types and formats

Engineering Contradiction:

Improveprocessing efficiencyVSAvoiddocument type coverage

Core Design Contradiction:

ProductivityVSAdaptability or versatility

Solution Approach 1:

The patent implements a universal font analysis system that can process various document types (driving licenses, passports, IDs) and formats by analyzing character fonts rather than relying on document-specific features like barcodes, making the system adaptable to diverse document standards across different countries

Inventive Principle:

Principle #6Universality (Multi-functionality)

Solution Approach 2:

The system changes the verification parameter from document-specific elements (barcodes, magnetic strips) to universal font characteristics that can be analyzed across different document types, allowing the same computational approach to handle varying document formats and anti-forgery measures

Inventive Principle:

Principle #35Parameter changes

3Reliability

If an automated system maintains an up-to-date database of document information to avoid false alerts, then assessment accuracy improves, but the device complexity and maintenance burden increase

Engineering Contradiction:

Improvefalse alert reductionVSAvoiddatabase maintenance

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The patent extracts and analyzes only the font characteristics from document images, separating this universal verification feature from the complex document-specific data that would require extensive databases, thereby simplifying the system while maintaining verification capability

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The font analysis system operates independently of external databases, using intrinsic font characteristics embedded in the document images themselves for verification, eliminating the need for continuous database updates and external information sources

Inventive Principle:

Principle #25Self-service

4Adaptability or versatility

If font analysis is performed on document images, then the method can handle various document formats, but imaging artefacts may reduce accuracy

Engineering Contradiction:

Improvedocument format handlingVSAvoidfont comparison accuracy

Core Design Contradiction:

Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The patent applies preliminary image processing steps including noise filtering, contrast enhancement, and artifact removal before font analysis, preparing the images in advance to ensure accurate font character extraction and comparison despite varying image quality

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system converts the challenge of imaging artefacts into an opportunity by using robust font feature extraction that focuses on stable, invariant characteristics of fonts that remain consistent despite image quality variations, turning potential sources of error into reliable verification signals

Inventive Principle:

Principle #22Blessing in disguise (Convert harm into benefit)

Data Source

PatentUS11087125B2Document authenticity determination

Publication Date: 2021.08.10 ONFIDO LTD

AI summary

A computer-implemented method for assessing if characters in a sample image are formed from a predefined font. The method comprises forming a first embedded space representation for the predefined font, extracting sample characters from the sample image, forming a second embedded space presentation of the sample characters, and comparing the first and second embedded space representation to assess if the sample characters are of the predefined font.