Access Control Head Detection Using Combined 2D and Depth Data

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current methods for verifying individuals at access control stations using two-dimensional portrait images are vulnerable to falsification through morphing or mask attacks, and capturing three-dimensional data is data-intensive and impractical.

Innovation Solution

A method that captures both a frontally aligned portrait image and a profile image of a person, combining these to generate depth information, which is compared with reference values stored on an identity document, enhancing security and reducing data requirements.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If three-dimensional data sets are captured and stored on an identity document chip, then security against morphing and mask attacks is improved, but hardware requirements and data intensity increase significantly making it impractical

Engineering Contradiction:
Improvesecurity against morphing and mask attacksVSAvoidhardware requirements and data intensity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent extracts only the essential depth information needed for security verification rather than capturing complete three-dimensional data sets. By taking out only the critical depth parameters from full 3D data, the system achieves security against morphing and mask attacks while avoiding the hardware complexity and data intensity of complete 3D capture systems

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent transitions from two-dimensional portrait images to incorporating depth information, effectively moving from 2D to a simplified 3D representation. This dimensional enhancement provides security against attacks that exploit 2D image vulnerabilities, while the selective extraction of depth data prevents the system from requiring full 3D capture hardware

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

2Quantity of substance

If only two-dimensional portrait photos are used for comparison, then data requirements are minimized, but the system becomes vulnerable to forgery through morphing or mask attacks

Engineering Contradiction:
Improvedata requirementsVSAvoidvulnerability to forgery
Core Design Contradiction:
Quantity of substanceVSReliability

Solution Approach 1:

The patent changes the parameters used for verification by adding depth information to the traditional 2D portrait photo parameters. This parameter expansion from purely 2D coordinates to include depth measurements increases reliability against forgery while maintaining reasonable data requirements through selective measurement of critical depth points

Inventive Principle:
Principle #35Parameter changes

3Reliability

If complete three-dimensional data sets are captured, then presentation attack detection is improved, but the data volume becomes excessively large for practical implementation

Engineering Contradiction:
Improvepresentation attack detectionVSAvoiddata volume
Core Design Contradiction:
ReliabilityVSQuantity of substance

Solution Approach 1:

The patent extracts only the essential depth information needed for security verification rather than capturing complete three-dimensional data sets. By taking out only the critical depth parameters from full 3D data, the system achieves security against morphing and mask attacks while avoiding the hardware complexity and data intensity of complete 3D capture systems

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent applies partial action by capturing only the specific depth information necessary for presentation attack detection rather than acquiring complete 3D data sets. This partial measurement approach provides sufficient security improvement while keeping data volumes manageable for practical implementation

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP3989183A1Method and device for optically detecting a head of a person to be checked at an access control station
Publication Date: 2022.04.27 BUNDESDRUCKEREI GMBH
  • EP3989183A1 patent drawingFigure 1~2
  • EP3989183A1 patent drawingFigure 3
  • EP3989183A1 patent drawing

AI summary

The invention relates to a method for optically capturing the head of a person (200) to be checked at an access control station (104), comprising the following steps: - Optically capturing a frontally oriented portrait image (114) of the head of the person (200) to be checked using a first camera (102) and obtaining a first two-dimensional data set from the portrait image (114), - Optically capturing a profile image (116) of at least one side of the head of the person (200) to be checked using the first camera (102) or using a second camera (112) and obtaining a second two-dimensional data set from the profile image (116), - Obtaining at least one depth information by combining the first data set with the second data set using an evaluation unit (108) of a processor unit (106),and - comparing the data sets and the depth information with reference values ​​contained on or stored in an identity document. The invention also relates to a method for the optical detection of a person to be checked (200). The invention also relates to an arrangement (100) for carrying out the method.