Mobile Iris Liveness Detection Using RGB-NIR Hyperspectral Imaging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Unsupervised authentication systems on mobile devices are susceptible to spoofing attacks, as they lack human supervision, making it easy for imposters to gain access using recorded or false biometric data, such as high-quality iris images or 3-D models.
Innovation Solution
Implement iris liveness detection using RGB and NIR image pairs captured synchronously by hybrid sensors, generating hyperspectral images, and analyzing pupil dynamics to distinguish between live and spoofed biometric data, employing classifiers and distance metrics to authenticate users.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If unsupervised authentication systems are implemented on mobile devices, then authentication convenience is improved, but security against spoofing attacks deteriorates
Solution Approach 1:
The system performs preliminary liveness detection by analyzing pupil dynamics and iris characteristics before completing authentication. This preliminary action detects spoofing attempts early in the authentication process, preventing unauthorized access while maintaining the convenience of automated authentication.
Solution Approach 2:
The system incorporates feedback mechanisms that analyze multiple characteristics (pupil response, iris texture, spectral properties) and use this feedback to dynamically adjust authentication decisions. The feedback loop enables the system to distinguish between live users and spoofing attempts, improving security without requiring manual supervision.
2Measurement precision
If liveness detection analysis is performed on captured images, then detection precision is improved, but processing time increases
Solution Approach 1:
The liveness detection process is segmented into multiple independent analysis stages: initial quality assessment, pupil dynamics analysis, iris texture analysis, and spectral analysis. This segmentation allows the system to process different characteristics in parallel and stop early if spoofing is detected, reducing overall processing time while maintaining high detection precision.
Solution Approach 2:
The system performs partial analysis by focusing on the most discriminative features (pupil response and iris texture) rather than analyzing all possible image characteristics. This partial action approach achieves sufficient detection precision with reduced computational overhead and faster processing.
3Measurement precision
If multi-spectral imaging is used for iris capture, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The mobile device's existing camera sensor is made multi-functional by capturing both visible spectrum images for normal photography and near-infrared images for iris authentication. This universality approach enables multi-spectral imaging without adding dedicated specialized hardware, maintaining device simplicity while improving measurement precision for liveness detection.
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 detects and prevents spoofing attacks on mobile devices, ensuring secure access by differentiating between genuine users and presentation attacks, even with advanced materials like 3-D models, without requiring special hardware or significant performance impact.
Implementation Method 1
A first classifier is applied to the trained model feature vectors and a feature vector generated for a hyperspectral image
Implementation Method 2
The mobile device may include a near infra-red (NIR) illuminator and a hybrid red-green-blue (RGB)/NIR sensor
Data Source
AI summary
An approach for an iris liveness detection is provided. A plurality of image pairs is acquired using one or more image sensors of a mobile device. A particular image pair is selected from the plurality of image pairs, and a hyperspectral image is generated for the particular image pair. Based on, at least in part, the hyperspectral image, a particular feature vector for the eye-iris region depicted in the particular image pair is generated, and one or more trained model feature vectors generated for facial features of a particular user of the device are retrieved. Based on, at least in part, the particular feature vector and the one or more trained model feature vectors, a distance metric is determined and compared with a threshold. If the distance metric exceeds the threshold, then a first message indicating that the plurality of image pairs fails to depict the particular user is generated. It is also determined whether at least one characteristic, of one or more characteristics determined for NIR images, changes from image-to-image by at least a second threshold. If so, then a second message is generated to indicate that the plurality of image pairs depicts the particular user of a mobile device. The second message may also indicate that an authentication of an owner to the mobile device was successful. Otherwise, a third message is generated to indicate that a presentation attack on the mobile device is in progress.