Rolling Shutter Illumination Synchronization for Biometric Spoof Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Biometric authentication systems face challenges in distinguishing between live individuals and two-dimensional spoof representations without using high-frame-rate cameras, which are expensive and complex.
Innovation Solution
The implementation of a rolling shutter mode in image capture devices synchronized with spatially separated illumination sources allows for higher frame rates and generation of 3D images using photometric stereo techniques, enabling effective spoof detection with lower-cost cameras.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a high-frame-rate camera is used to capture images for 3D representation and spoof detection, then the reliability of biometric authentication is improved, but the device complexity and cost increase
Solution Approach 1:
The patent divides the image capture process into multiple sequential passes using a rolling shutter camera. Different sets of pixels (e.g., odd rows and even rows) are captured in alternating frames under different illumination conditions. This segmentation allows a low-frame-rate camera to effectively capture multiple images per frame interval, achieving the same spoof detection capability as high-frame-rate cameras without the associated complexity and cost.
Solution Approach 2:
The patent employs periodic illumination sequences where illumination sources are alternately activated in synchronization with the rolling shutter camera's pixel row exposure. This periodic action creates alternating illumination patterns across different pixel sets, enabling the capture of multiple photometric stereo images sequentially. The periodic modulation of illumination combined with the rolling shutter scan effectively multiplies the temporal sampling rate without requiring a higher camera frame rate.
2Productivity
If multiple images are captured sequentially with a rolling shutter camera, then the frame rate is effectively increased for 3D generation, but the resolution of individual images decreases
Solution Approach 1:
The patent segments the image sensor into multiple sets of pixels (e.g., different row groups) that are exposed at different times during the rolling shutter sequence. Each set captures a portion of the full image, and these partial images are later combined or processed separately. This segmentation enables the system to capture multiple temporal samples (increasing effective frame rate) while each individual pixel set maintains sufficient spatial resolution for its purpose.
Solution Approach 2:
The patent transitions from capturing complete images at a single moment to capturing spatial segments across multiple time dimensions. By utilizing the temporal dimension of the rolling shutter scan and assigning different spatial pixel sets to different illumination conditions and time points, the system effectively adds a time dimension to the image capture process. This dimensional transformation allows multiple images to be synthesized from sequential pixel data while preserving sufficient resolution in each synthesized image for 3D reconstruction.
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 the reliability and security of biometric authentication by differentiating between live persons and 2D spoof representations, reducing hardware requirements and costs.
Implementation Method 1
A three-dimensional image is generated using the data corresponding to the first set of pixels as a first image of a pair of photometric stereo images, and the data corresponding to the second set of pixels as a second image of the pair of photometric stereo images
Data Source
AI summary
Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for a biometric authentication system. In one aspect, a method includes receiving, at one or more processing devices, data corresponding to a first set of pixels of an image sensor. The first set of pixels are exposed under illumination by a first source. Data corresponding to a second set of pixels of the image sensor is received, at the one or more processing devices. The second set of pixels are exposed under illumination by a second source that is spatially separated from the first source. A three-dimensional image is generated using the data corresponding to the first set of pixels as a first image of a pair of photometric stereo images, and the data corresponding to the second set of pixels as a second image of the pair of photometric stereo images.