Biometric Identification Using Subset Indexing for Faster Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing biometric identification systems face challenges in efficiently identifying individuals from large populations, leading to long processing times and high computing resource demands, while relying on civil status filters introduces fraud risks.
Innovation Solution
A biometric identification method that selects a derived population subset by intersecting multiple sets of target individuals based on biometric classification criteria, reducing the comparison scope to a smaller, reliable subset for faster and more secure identification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the biometric characteristics of the individual to be identified are compared with the biometric characteristics of the entire predetermined population, then the reliability of identification is improved, but the identification time increases significantly
Solution Approach 1:
The predetermined population is divided into multiple subsets based on biometric classification criteria (e.g., age, gender, ethnicity). Instead of comparing against the entire population, the system segments the search space into manageable groups, reducing the number of comparisons required while maintaining comprehensive coverage through systematic subset evaluation.
Solution Approach 2:
Biometric classification criteria are applied in advance to pre-categorize the population into distinct subsets. This preliminary classification allows the identification process to start with smaller, pre-grouped populations rather than the full population, significantly reducing comparison time while ensuring all relevant groups are systematically evaluated.
2Productivity
If the number of computing units is increased to accelerate identification, then the identification speed is improved, but the hardware cost and complexity increase
Solution Approach 1:
The population data is segmented into multiple subsets stored in the database, allowing a single computing unit to process comparisons more efficiently by working with smaller subsets rather than the entire population. This reduces the computational burden on each unit while maintaining overall identification speed.
Solution Approach 2:
The system performs partial comparisons by evaluating biometric characteristics against multiple subsets sequentially or in parallel, rather than requiring all computing units to process the entire population simultaneously. This approach achieves adequate identification speed with fewer hardware resources.
3Loss of time
If civil status filters are used to reduce the population being compared, then the identification time is reduced, but the risk of fraud increases
Solution Approach 1:
The system changes the classification parameters from civil status data (which can be falsified) to biometric-based criteria such as physical characteristics, demographic attributes derived from biometric analysis, or other objective biological parameters. These parameters are more difficult to manipulate and provide a more reliable basis for population segmentation.
Solution Approach 2:
Biometric classification criteria serve as an intermediary between raw biometric data and the final identification comparison. This intermediary layer processes the biometric characteristics to determine appropriate subsets for comparison, providing an objective, fraud-resistant mechanism for reducing the population scope without compromising security.
Data Source
Figure 1
Figure 2a
Figure 2b
AI summary
The invention relates to a method for identifying a person and comprising: a step of acquiring a captured visual or audio input from said person comprising a plurality of biometric characteristics; a step of extracting the biometric characteristics; a step of determining at least one biometric classification criterion; a step of indexing said input to at least one list chosen from a plurality of pre-loaded lists according to a classification criterion; a step of comparing one or more extracted biometric characteristics with the pointed biometric characteristics of a restricted subset of the population.