Graphometric Signature Recognition via Fragmented Biometric Encoding

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing systems for managing digital graphometric signatures face security vulnerabilities due to the potential for unauthorized access to stored data and dependency on specific acquisition devices with varying sensor capabilities, which can compromise the safety and efficiency of signature recognition.

Innovation Solution

A system and procedure for managing digital signatures that securely registers and recognizes signatures by discretizing them into fragments with associated biometric data, translating this data into symbolic reference codes stored on a secure server, making it difficult to trace the original signature and ensuring compatibility across a range of electronic devices.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If digital signatures are stored on archiving servers with detailed node characteristics, then signature recognition accuracy is improved, but security against unauthorized access and hacking is worsened

Engineering Contradiction:
Improvesignature recognition accuracyVSAvoidsecurity vulnerability to hacking
Core Design Contradiction:
Measurement precisionVSObject-affected harmful factors

Solution Approach 1:

The patent divides the digital signature into multiple discrete nodes with specific coordinates and characteristics. Each node is processed and stored separately, allowing the system to reconstruct and recognize the signature without storing the complete signature image. This segmentation enables accurate recognition while reducing security risks associated with storing complete signature data.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent extracts only the essential biometric parameters from the signature (node coordinates, arrival times, pressure values) and stores these extracted features rather than the complete signature. This extraction process removes unnecessary information that could be used for tracing or hacking while retaining sufficient data for accurate recognition.

Inventive Principle:
Principle #2Taking out (Extraction)

2Reliability

If signature acquisition systems use multiple sensor types for comprehensive biometric data collection, then recognition reliability is improved, but device complexity and compatibility issues are worsened

Engineering Contradiction:
Improverecognition reliabilityVSAvoidsensor requirements
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent designs the signature acquisition system to work with multiple types of devices (smartphones, tablets, graphic tablets, computers) by identifying and utilizing biometric parameters that can be detected by various sensor types. The system accommodates different sensor capabilities while maintaining consistent recognition performance across diverse devices.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The patent adapts the signature recognition process to work with different sensor capabilities by adjusting which biometric parameters are collected and how they are processed. The system can operate with varying levels of sensor precision and types, changing the parameter set based on device availability while maintaining recognition reliability.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP4002202A1Recognition system and management procedure for graphometric digital signatures
Publication Date: 2022.05.25 TENSORLOOPS SRL
  • EP4002202A1 patent drawingFigure 1
  • EP4002202A1 patent drawingFigure 2
  • EP4002202A1 patent drawing

AI summary

It is provided a procedure (I) for managing a digital signature comprising a registration step of at least a first handwritten signature (11') of a user including acquiring (II) from the user a plurality of first digital signatures (12) derived from first handwritten signatures (11) and a plurality of first biometric data (13) associated with the first handwritten signatures (11); execute (III) a computer program configured to compare the first digital signatures (12) and discretize the first digital signatures (12) into a plurality of first fragments (12a) to which one or more first biometric data are associated with each one (13 ) recurring among the first digital signatures (12) within predetermined tolerances; correlating (IV), through the program, each first fragment (12a) to respective one or more first biometric data (13) defining first correlated data (14); sequentially translate (V) the first correlated data (14) into a reference code (15) on the basis of a predetermined logic in such a way as to define digital data (10), recording (VI) the digital data (10) on a first storage support (2) in such a way that the first related data (14) cannot be viewed starting from the first support (2), associating (VII) on the first support (2) an identification code (ID) of said user to the digital data (10).