Living Skin Tissue Tracking via Spectral Reflectance Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current facial recognition systems face challenges in distinguishing between live human skin tissue and image representations, leading to potential spoof attacks, which compromises the security of biometric authentication methods.
Innovation Solution
A system and method that utilize a machine learning classifier trained on spectral reflectance features extracted from video sequences to differentiate between living skin tissue and non-living representations by analyzing regions of interest (ROI) in video frames, employing techniques such as Fourier transforms and Independent Component Analysis (ICA) to determine the presence of living skin tissue.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If traditional facial recognition systems are used, then authentication speed is improved, but security against spoof attacks deteriorates
Solution Approach 1:
The patent changes the detection parameters from simple facial feature recognition to spectral reflectance analysis across multiple wavelengths. By measuring how living skin tissue reflects different wavelengths of light compared to non-living materials, the system maintains fast authentication while adding a security dimension that detects physiological properties unique to living tissue.
Solution Approach 2:
The patent adds a new dimension of analysis by incorporating spectral reflectance measurements across multiple wavelengths alongside traditional facial recognition. This multi-dimensional approach combines speed of traditional methods with the security of physiological detection, allowing the system to authenticate quickly while verifying liveness through optical properties.
2Measurement precision
If spectral reflectance analysis is performed on video sequences, then detection accuracy of living skin tissue is improved, but computational complexity increases
Solution Approach 1:
The patent segments the analysis by first identifying regions of interest (facial areas) in video frames, then applying spectral reflectance analysis only to those specific regions rather than processing entire video sequences. This segmentation reduces computational complexity while maintaining detection accuracy by focusing resources on relevant areas.
Solution Approach 2:
The patent performs preliminary facial region detection and segmentation before applying the computationally intensive spectral reflectance analysis. By pre-identifying and isolating facial regions of interest, the system reduces the data volume requiring complex analysis, thereby lowering computational complexity while preserving measurement precision on the critical areas.
3Measurement precision
If multiple features from video frames are extracted, then classification accuracy is improved, but processing time increases
Solution Approach 1:
The patent extracts only the most discriminative spectral reflectance features from video frames rather than processing all available data. By selecting and extracting key features that most effectively distinguish living skin tissue from non-living materials, the system achieves high classification accuracy while minimizing processing time through focused feature extraction.
Data Source
AI summary
A method comprising: receiving, as input, a video sequence comprising a plurality of frames and depicting a scene; detecting, in each of said frames, one or more regions of interest (ROI) associated with an object in said scene; extracting, from each of said ROIs with respect to all of said frames, a feature set representing spectral reflectance; at a training stage, train a machine learning classifier on a training set comprising: (i) all of said feature sets, and (ii) labels indicating whether each of said ROIs depicts living skin tissue; and at an inference stage, applying said trained machine learning classifier to a target feature set extracted from a target ROI detected in a target video sequence, to determine whether said ROI depicts living skin tissue.