Dual-Focal Face Recognition Fraud Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Biometric face recognition systems are vulnerable to fraud attempts, such as displaying a digital image of an authorized person's face on a screen, which can deceive the system by making the screen edges invisible and exploiting the limited field of view of optical sensors.
Innovation Solution
The method employs two image sensors with different field angles, one with a wider angle to detect screen borders and the other with a narrower angle to detect moiré effects, ensuring simultaneous image capture to verify the authenticity of the face presented, thereby preventing fraudulent access.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a single optical sensor with narrow field of view is used for face recognition, then the recognition precision is improved, but the system becomes vulnerable to fraud using displayed images
Solution Approach 1:
The system divides the detection task into two segments: one sensor (first sensor) detects screen borders around the face, while the other sensor (second sensor) detects moiré effects. This segmentation allows each sensor to specialize in detecting different fraud indicators, thereby improving overall reliability without sacrificing recognition precision
Solution Approach 2:
The patent introduces an intermediary analysis process that examines images from both sensors for specific fraud indicators (borders and moiré effects). This intermediary detection layer acts as a mediator between the image capture and final recognition decision, providing an additional reliability check that doesn't interfere with the precision of the primary recognition function
2Measurement precision
If the field of view is narrowed to focus on the face, then the recognition accuracy is improved, but the ability to detect screen borders is reduced
Solution Approach 1:
The detection function is segmented between two sensors: the first sensor with wider field of view captures the broader area including potential screen borders, while the second sensor with narrower field of view focuses on the face region for accurate recognition and moiré detection. This segmentation resolves the contradiction by assigning different spatial coverage responsibilities to different sensors
Solution Approach 2:
The system transitions from a single-dimension (single sensor) approach to a two-dimension (dual sensor with different field angles) approach. By adding the dimensional aspect of varying field angles, the system can simultaneously achieve both wide area coverage for border detection and narrow focused coverage for accurate face recognition
3Reliability
If images are captured simultaneously with different field angles, then fraud detection capability is improved, but the device complexity increases
Solution Approach 1:
Both image sensors are designed with multi-functionality: they can detect faces for recognition, detect screen borders, and detect moiré effects. This universality reduces the need for entirely separate detection systems, thereby limiting the increase in device complexity while maintaining enhanced fraud detection capability
Solution Approach 2:
The patent merges the fraud detection function with the existing face recognition imaging system. Instead of adding completely separate detection devices, the system combines border detection and moiré detection capabilities into the existing dual-sensor imaging architecture, thereby achieving enhanced reliability with minimal additional complexity
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 effectively secures automatic face recognition by distinguishing between real and displayed images, enhancing the system's ability to detect and prevent unauthorized access.
Implementation Method 1
acquiring at least one first image of the face by means of a first sensor having a first field angle
Implementation Method 2
analyzing the second image to verify that there is no moiré effect; and detecting the absence of fraud when there is no frame and no moiré effect
Data Source
AI summary
A method and an associated device for detecting fraud during automatic face recognition, the method comprising the following steps: acquiring a first image of the face by means of a first sensor having a first field angle, and a second image of the face by means of a second sensor having a second field angle that is narrower than the first field angle; analyzing the first image to verify that there is no frame around the face; and analyzing the second image to verify that there is no moiré effect.