Spatially Patterned Illumination for Fingerprint Presentation Attack Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Fingerprint scanners face challenges in detecting presentation attacks, where fake fingerprints are used to evade identification, as existing technologies lack effective methods to differentiate between real human skin and spoofed versions.
Innovation Solution
A fingerprint scanning system that uses spatially patterned illumination to analyze the interaction of light with the platen surface, employing a combination of optical sensing and machine learning algorithms to determine whether the object is human skin or a presentation attack, by capturing images with varying light patterns and analyzing color absorption and scatter properties.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional fingerprint sensing is used, then identification and verification functions are provided, but the system cannot detect presentation attacks or fake fingerprints
Solution Approach 1:
The patent combines multiple sensing modalities (optical imaging, capacitive sensing, and spectral analysis) into a unified fingerprint authentication system. The optical imaging module captures reflectance images, the capacitive sensor detects electrical properties, and the spectral analyzer measures light absorption characteristics, merging these functions to achieve both traditional identification and presentation attack detection without requiring separate systems
Solution Approach 2:
The patent introduces spectral analysis as an intermediary detection layer that analyzes the optical properties of the fingerprint sample. By measuring the absorption spectrum at multiple wavelengths, the system can identify characteristic signatures of real skin versus fake materials, serving as a mediator between the physical fingerprint sample and the authentication decision
2Adaptability or versatility
If fake fingerprints are used, then impersonation and evasion of identification are achieved, but the system lacks the capability to differentiate real human skin from spoofed versions
Solution Approach 1:
The patent changes the detection parameters from single-wavelength optical sensing to multi-wavelength spectral analysis. By measuring absorption characteristics across multiple wavelengths (including visible and near-infrared ranges), the system captures unique spectral signatures that differentiate real human skin from fake fingerprint materials, making the detection parameter more discriminative
Solution Approach 2:
The patent employs a multi-modal sensing approach that combines multiple types of sensors (optical, capacitive, spectral) to create a composite detection system. This composite approach leverages the complementary strengths of each sensing modality, where optical imaging provides structural information, capacitive sensing detects electrical properties, and spectral analysis identifies material composition, together forming a robust system that can distinguish real from fake fingerprints
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Effectively differentiates between real human skin and spoofed fingerprints, enhancing security by accurately identifying genuine human fingerprints and detecting presentation attacks, thereby improving the reliability of fingerprint-based identification systems.
Implementation Method 1
an illumination module configured to illuminate at least a portion of the object with light from an illumination source
Implementation Method 2
analyzing color absorption and scatter properties
Implementation Method 3
analyzing color absorption and scatter properties
Data Source
AI summary
An apparatus for capturing an image of an object on a platen and including a platen having a surface for receiving the object, wherein the platen is transmissive to an optical wavelength of light, an illumination module configured to illuminate at least a portion of the object with light from an illumination source, and an optical sensing module configured to receive the light from the illumination source after the light interacts with the at least a portion of the object. The light from the illumination source is spatially patterned prior to reaching the object. The illumination module can be configured to create the spatially patterned illumination and/or the platen may include a patterned optical coating creating the spatially patterned illumination. The spatially patterned illumination can include a plurality of discrete areas (e.g., a stepped pattern) of different illumination intensity and/or a gradient of different illumination intensity.