Liveness Detection via Projected Light Reflection Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current face recognition systems are vulnerable to spoofing attacks, as they often require user interaction and may not effectively differentiate between live and spoofed faces, especially after image normalization or transformation processes that can filter out spoofing features.
Innovation Solution
The system uses a projector to display varying colors, patterns, and sequences of images onto the subject's face and analyzes the reflected images to determine whether the subject is live or a spoof, utilizing software to differentiate between 3D live subjects and 2D images or videos.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional face recognition systems are used, then the system is simple and easy to operate, but it is vulnerable to spoofing attacks and cannot effectively differentiate between live and spoofed faces
Solution Approach 1:
The system segments the face recognition process into multiple independent stages: liveness detection stage (using reflected image analysis) and face recognition stage (using normalized image features). This segmentation allows the liveness detection mechanism to operate independently without complicating the overall face recognition system, while effectively detecting spoofing attempts at an early stage.
Solution Approach 2:
The system performs liveness detection as a preliminary action before conducting face recognition. By analyzing reflected images and detecting user interaction behaviors (such as blinking, head movement, or eye movement) in advance, the system can filter out spoofed attempts before the main recognition process, thereby improving security without adding complexity to the core recognition algorithm.
2Measurement precision
If image normalization or transformation processes are applied, then the face recognition accuracy is improved, but spoofing features are filtered out
Solution Approach 1:
The system separates the processing pipeline into two distinct segments: one for liveness detection using reflected images with minimal processing, and another for face recognition using normalized images. This segmentation preserves spoofing features in the reflected image analysis while applying normalization only to the recognition stage, thus maintaining both accuracy and liveness detection capability.
Solution Approach 2:
The system uses reflected images as an intermediary medium to detect liveness before the main face recognition process. The reflected images capture real-time light interactions and user behaviors that serve as intermediaries to verify liveness, while the subsequent normalized images are used solely for recognition without compromising the liveness information captured in the reflected images.
3Reliability
If user interaction is required for liveness detection, then the detection accuracy is improved, but the user experience becomes less convenient
Solution Approach 1:
The system enables self-service liveness detection by automatically analyzing reflected images for user interaction behaviors without requiring explicit user actions or awareness. The system detects natural behaviors such as blinking, head movement, and eye movement that occur during normal use, allowing the device to verify liveness autonomously while maintaining good user experience.
Solution Approach 2:
The system implements continuous feedback by monitoring reflected images in real-time to detect user interaction behaviors. The analysis of changes in reflected light patterns provides feedback about user liveness status, enabling the system to adapt and maintain secure authentication without interrupting user interaction or requiring additional user actions.
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
This approach enhances liveness detection by providing a more convenient user experience and effectively differentiating between live and spoofed faces at earlier stages of the enrollment or recognition process, reducing the risk of false access attempts.
Implementation Method 1
Analysis of reflections of projected light in varying colors, brightness, patterns, and sequences for liveness detection in biometric systems
Data Source
AI summary
Spoofed faces are unlikely to reflect a liveness detection image in the same manner as live faces. To determine liveness of a subject face, a liveness detection image is displayed on a screen of a face recognition device in the direction of the subject face, and a reflection image of the subject face is captured while the liveness detection image is displayed. The reflection image is analyzed to determine whether the reflection image contains any reflections of the liveness detection image, and a liveness determination is made based on the location and/or character of any such reflections.