Multimodal Identity Verification Using Facial and Eye-Print Analysis

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current user identity verification methods, relying on facial image recognition, face challenges with low precision and reliability due to advancements in synthesizing lifelike 3D facial images and simulated movements, compromising security in financial transactions.

Innovation Solution

Implementing a user identity verification system that combines facial verification with eye-print image verification, using multidimensional verification modes, where a facial quality score threshold and eye-print collection steps are set to obtain and compare facial and eye-print images against presets, enhancing precision and reliability.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If facial image recognition with live facial image verification is used, then the verification process is simple and fast, but the precision and reliability are low due to synthesizable 3D facial images

Engineering Contradiction:
Improveidentity verification reliabilityVSAvoidverification system complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent divides the facial verification process into multiple independent components: facial image verification, eye-print image verification, and live eye-print verification. Each component focuses on a specific region or aspect of the face, making the overall system more reliable while maintaining manageable complexity through modular architecture

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent transitions from traditional 2D facial recognition to 3D spatial verification by capturing eye-print images at multiple collection steps with different spatial coordinates. This adds a dimensional aspect to verification, making it resistant to 2D synthesized images while maintaining practical implementability

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Measurement precision

If traditional facial recognition is used, then the operation is simple, but the measurement precision is low due to inability to distinguish synthesized images

Engineering Contradiction:
Improveidentity verification precisionVSAvoidverification time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The system performs preliminary actions by collecting multiple eye-print images at different collection steps before final verification. This pre-collection of verification data from multiple angles and timestamps creates a robust baseline that enables precise verification while streamlining the actual verification process

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent uses multiple temporary eye-print images captured at different collection steps as disposable verification elements. Each eye-print image serves a specific verification purpose and is then discarded, allowing the system to achieve high precision through multiple low-cost verification attempts rather than relying on a single complex verification step

Inventive Principle:
Principle #27Cheap short-living objects (Disposable)

Data Source

PatentEP3506589B1User identity verification method, apparatus and system
Publication Date: 2021.06.23 ADVANCED NEW TECHNOLOGIES CO LTD
  • EP3506589B1 patent drawingFigure 1~2
  • EP3506589B1 patent drawingFigure 3~5
  • EP3506589B1 patent drawingFigure 6~7

AI summary

This invention discloses a user identity verification method, apparatus, and system, relating to the field of information technology. This invention primarily is used to solve the problems of low precision and reliability in current user identity verification methods. The method comprises: first receiving a facial image and one or more eye-print pair images corresponding to an identity verification object from a client, the one or more eye-print pair images corresponding to a number of eye-print collection steps, then comparing the facial image to a preset facial image, comparing the eye-print pair images to preset eye-print templates, and sending successful identity verification information to the client if the comparison results for the facial image and the eye-print pair images meet preset conditions.