Biometric Identification Using Robust Principal Component Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current biometric identification systems face challenges in accurately and securely identifying individuals using machine learning techniques, particularly in removing noise from biometric measurement signals and determining matches efficiently.
Innovation Solution
The method involves receiving a biometric measurement signal, extracting periodic fragments, generating feature data, removing noisy data using robust principal component analysis (RPCA), and using machine learning techniques, such as deep learning, to determine matches between the processed data and known biometric data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If machine learning techniques are used to process biometric measurement signals, then identification accuracy is improved, but noise in the biometric data affects matching reliability
Solution Approach 1:
The patent extracts periodic fragments from the biometric measurement signal to isolate useful information from noise. This extraction process separates the meaningful periodic patterns from the noisy background, allowing accurate identification while maintaining matching reliability despite noise presence in the original signal
Solution Approach 2:
The patent performs preliminary processing steps including extracting periodic fragments and generating feature data before conducting the actual matching operation. This preliminary action prepares clean, processed data for comparison, ensuring that the matching process works with refined information rather than raw noisy data
2Reliability
If complex processing steps are applied to remove noise and extract features, then identification reliability is improved, but processing time increases
Solution Approach 1:
The patent segments the biometric measurement signal into periodic fragments, processing them in discrete units. This segmentation allows efficient extraction of features from each fragment and enables parallel or sequential processing that maintains reliability while managing processing time through structured breakdown of the signal
Solution Approach 2:
The patent transforms the biometric measurement signal into feature data by changing parameters such as extracting periodicity characteristics and converting time-domain signals into frequency-domain representations. This parameter transformation reduces data complexity while preserving identification-relevant information, improving reliability without excessive processing time
Data Source
AI summary
Method and apparatus for processing a biometric measurement signal using a computing device, including receiving a biometric measurement signal generated by contact with a single individual, extracting at least one periodic fragment from the biometric measurement signal, generating first feature data at least partially based on the at least one extracted periodic fragment, determining second feature data from the first feature data by removing data from the first feature data using robust principal component analysis, determining whether a match exists between the second feature data and defined biometric data associated with a known individual by processing the second feature data and the defined biometric data using a machine learning technique, and in response to determining a match exists between the second feature data and the defined biometric data, transmitting a signal indicating that the single individual is the known individual.