Forged Face Detection Using Infrared and Photorealistic Image Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional biometric face recognition systems face challenges in distinguishing actual from forged biological information, and the use of thermal infrared cameras is expensive and inconvenient for users.
Innovation Solution
A method and apparatus that simultaneously use an infrared camera and a still camera to capture face images, analyze the characteristics of the infrared image to determine if a face is forged, and perform user authentication using a photorealistic image if the face is not forged, incorporating image processing techniques like Adaboost and SVM for face region extraction and quality evaluation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If thermal infrared camera is used to capture face images, then ability to distinguish actual face from forged face is improved, but equipment cost increases
Solution Approach 1:
The patent combines a standard visible light camera and a thermal infrared camera into a single imaging system. The visible light camera captures photorealistic images while the thermal infrared camera captures thermal distribution images, and both are processed together to enhance forged face detection capability without requiring expensive specialized equipment alone
Solution Approach 2:
The patent introduces an intermediary processing system that includes quality characteristic extraction units and forged face determination units. These intermediaries analyze the thermal distribution patterns and compare them with photorealistic images to detect forged faces, serving as a bridge between the captured images and the authentication decision
2Reliability
If thermal infrared camera is used to capture face images, then ability to distinguish actual face from forged face is improved, but user convenience deteriorates
Solution Approach 1:
By merging visible light and thermal infrared imaging capabilities into one system, the patent enables simultaneous capture of both image types without requiring users to switch between different devices or perform separate actions, thus maintaining user convenience while improving detection reliability
Solution Approach 2:
The system automatically performs quality characteristic extraction and forged face determination without requiring user intervention. The processing units autonomously analyze thermal distribution patterns and make authentication decisions, eliminating the need for users to manually adjust settings or provide additional information
3Device complexity
If conventional biometric recognition algorithm is used, then authentication process is simple, but ability to distinguish actual biological information from forged biological information deteriorates
Solution Approach 1:
The patent segments the authentication process into distinct functional units: image acquisition units for capturing visible light and thermal infrared images, quality characteristic extraction units for analyzing thermal distribution patterns, and forged face determination units for making authentication decisions. This segmentation allows complex analysis to be performed through modular, manageable processing stages
Solution Approach 2:
The patent replaces simple conventional biometric algorithms with a multi-stage processing system that incorporates thermal infrared image analysis. Instead of relying solely on standard image processing, the system uses thermal distribution characteristics as an additional dimension for authentication, substituting basic mechanical recognition with more sophisticated thermal-based verification
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 improves the reliability and security of face detection by effectively distinguishing actual from forged faces without the need for expensive thermal infrared cameras and reduces user inconvenience.
Implementation Method 1
acquiring an infrared image and a photorealistic image captured by one camera or two cameras
Data Source
AI summary
A method for detecting a forged face using an infrared image includes: acquiring an infrared image and a photorealistic image captured by one or two cameras; extracting a face region from the photorealistic image; determining, based on analysis of quality characteristics of the infrared image, whether a current face is a forged face or not; and performing face recognition on the extracted face region if it is determined that the current face is a non-forged face. The method further includes capturing an infrared image and a photorealistic image again through the camera without performing face recognition if it is determined that the current face is a forged face.