Biometric Template Matrix Selection for Memory Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The challenge is to efficiently store and match biometric data templates collected using small biometric sensors, which often result in insufficient overlap between earlier and new data, leading to complexity in template selection and increased memory requirements.
Innovation Solution
A method is developed to align biometric data templates, assign a matrix of cells to determine the information content, and select a limited set of templates based on distinguishable data cells that meet specific criteria, such as maximum cell count or quality parameters, to ensure effective storage and matching.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a large amount of biometric templates is stored to increase overlap likelihood, then matching reliability is improved, but memory space requirement increases and template selection complexity increases
Solution Approach 1:
The patent divides the biometric template selection process into segments by introducing a matrix of cells that categorizes templates based on their information content and distinguishability. Instead of treating all templates equally, the system segments them into cells representing different regions or characteristics of the biometric object, allowing for structured selection that reduces complexity while maintaining reliability.
Solution Approach 2:
The patent changes the parameter for template selection from arbitrary or random selection to selection based on information content metrics. By evaluating templates according to their distinguishable data content and assigning them to matrix cells based on these parameters, the system enables intelligent selection of a limited set of templates that maximizes matching reliability without requiring storage of all possible templates.
2Reliability
If multiple biometric readings are used for the template to increase overlap likelihood, then matching reliability is improved, but memory space requirement and processing complexity increase
Solution Approach 1:
The patent extracts only the essential and distinguishable information from multiple biometric readings by organizing template data into a matrix of cells. Instead of storing complete redundant copies of multiple readings, the system extracts unique information content from each reading and places it in appropriate cells, thereby reducing memory space requirements while preserving the reliability benefits of multiple readings.
Solution Approach 2:
The patent merges information from multiple biometric readings into a unified matrix structure where overlapping information is consolidated. By combining data from multiple readings into shared cells rather than storing them separately, the system achieves the reliability of multiple readings while minimizing memory space through efficient data consolidation.
3Quantity of substance
If a limited set of templates is stored to reduce memory space, then memory efficiency is improved, but matching reliability may deteriorate due to insufficient overlap
Solution Approach 1:
The patent applies local quality by ensuring that each cell in the matrix contains templates with specific local characteristics or information content. By optimizing the quality and distinguishability of templates within each local cell rather than treating all templates uniformly, the system achieves high matching reliability with a limited number of templates, as each template contributes unique local information to the overall matching capability.
Data Source
AI summary
There is provided a method of storing a set of biometric data templates, which emanates from a biometric object, among a plurality of biometric data templates, wherein each biometric data template comprises biometric data from only a part of a biometric object. The method comprises aligning biometric data templates with each other in view of the biometric object of the plurality of biometric data templates which comprises partially overlapping biometric data with at least one other biometric data template; determining the set of biometric data templates to be stored; and storing the set of biometric data templates. The determining of the set of biometric data templates to be stored comprises assigning a matrix of cells covering at least partially the aligned biometric data templates; determining, for each of the aligned biometric data templates, cells comprising distinguishable biometric data; and determining a combination of biometric data templates which comprises an amount of cells comprising distinguishable biometric data which fulfil a predetermined criterion. A method of biometric matching is also disclosed, and also a biometric matching apparatus and a computer program.