Liveness Testing via Diffusion Analysis for Face Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional face recognition systems are vulnerable to impersonation using pictures, as they fail to distinguish between real and fake images effectively.
Innovation Solution
A liveness testing method that involves extracting features from images by analyzing diffusion speeds and light energy distribution to determine if an object is a 3D real face or a 2D representation, using a diffusion equation and statistical analysis to differentiate between the two.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional face recognition systems are used, then user recognition is simple and fast, but the system is vulnerable to impersonation using pictures
Solution Approach 1:
The patent performs a liveness test before conducting face recognition to determine whether the captured image represents a real person or a photograph. This preliminary action prevents impersonation attacks by verifying the authenticity of the subject before proceeding with recognition, thereby resolving the contradiction between security and system simplicity.
Solution Approach 2:
The patent divides the face recognition system into two distinct modules: a liveness test module that verifies whether the subject is real or a photograph, and a face recognition module that performs identification. This segmentation allows each module to specialize in its function, improving overall reliability while maintaining manageable system complexity.
2Reliability
If liveness testing with diffusion analysis is performed, then impersonation is prevented, but processing time and computational complexity increase
Solution Approach 1:
The patent applies diffusion analysis selectively to specific regions of the face image rather than the entire image, and performs the liveness test only when needed (e.g., when recognition confidence is below a threshold). This partial application reduces processing time while maintaining detection accuracy, resolving the contradiction between reliability and processing speed.
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 prevents impersonation by accurately distinguishing between real and fake images, enhancing the security of face recognition systems.
Implementation Method 1
generating a second image by diffusing a plurality of pixels included in the first image; calculating diffusion speeds of the pixels based on a difference between the first image and the second image
Implementation Method 2
estimating a surface property related to an object included in the first image based on the diffusion speeds. The surface property may include at least one of a light-reflective property of a surface of the object
Data Source
AI summary
A user recognition method and apparatus, the user recognition method including performing a liveness test by extracting a first feature of a first image acquired by capturing a user, and recognizing the user by extracting a second feature of the first image based on a result of the liveness test, is provided.