Biometric Hashing via Minutiae Extraction and Salted Verification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Biometric data, particularly fingerprint scans, are prone to inexactness due to variations in scanning positions and orientations, making them vulnerable to spoofing and requiring large storage and transmission, which is inefficient over slower networks, and cannot be hashed for secure comparison.
Innovation Solution
A system that generates a digital biometric string by identifying central features and minutiae in fingerprint scans, combining them with cryptographic salts to create hash signatures that can be verified locally without storing or transmitting the actual scan, allowing for secure and efficient authentication.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If biometric scan data is stored on servers for verification, then authentication can be performed, but the data becomes vulnerable to server attacks and hacking
Solution Approach 1:
The patent extracts only the essential verification information (minutiae points and their relative positions) from the complete biometric scan, storing only this extracted data on servers rather than the full scan. This allows authentication to proceed while minimizing the vulnerability of stored data, as the extracted features cannot be reverse-engineered to recreate the original biometric image.
Solution Approach 2:
The patent creates a simplified copy or representation of the biometric data (a set of numerical values describing minutiae positions) that serves the authentication function without containing the actual sensitive biometric information. This copy is what gets stored and transmitted, replacing the need to store the original scan image.
2Reliability
If complete biometric scan data is transmitted over networks, then verification can be performed, but transmission time increases significantly on slower networks
Solution Approach 1:
The system extracts only the critical verification elements (minutiae positions and relationships) from the complete biometric scan, creating a compact data representation that is much smaller than the original scan. This extracted data can be transmitted quickly over networks while still providing sufficient information for accurate verification.
3Object-affected harmful factors
If biometric scan data is hashed prior to transmission, then security is improved, but inexact scans cause hash comparison to fail
Solution Approach 1:
The patent performs preliminary processing of the biometric scan to extract stable, consistent features (minutiae points and their relative positions) before hashing. By preprocessing the data to identify these invariant characteristics, the system ensures that even if the overall scan varies slightly, the extracted features remain consistent and produce reliable hash comparisons.
4Measurement precision
If fingerprint scan variations are accommodated by storing multiple scans, then matching accuracy improves, but data storage requirements increase
Solution Approach 1:
Instead of storing multiple complete scan images to accommodate variations, the system extracts the essential verification features (minutiae positions and relationships) that remain consistent across different scans. This extracted representation is stored once and is sufficient for accurate matching, dramatically reducing storage requirements while maintaining matching accuracy.
Data Source
AI summary
A fingerprint identification system comprising a smart device, a fingerprint scanner, a processor coupled to a transceiver and to the fingerprint scanner, and a digital storage element coupled to the processor. The digital storage element stores logic that causes the processor to: activate the fingerprint scanner to scan a user's finger, identify a feature and multiple minutia of the user's fingerprint, and generate a digital fingerprint string(s) indicative of a position of each minutia relative to the feature. The processor is then caused to combine the digital fingerprint string(s) with a first cryptographic salt to generate a first hash. The first hash is compared to a first hash signature to determine if the first hash represents an authentic fingerprint. If authentic, the processor combines the digital fingerprint string(s) with a second cryptographic salt to generate a second hash. This second hash is transmitted to one or more servers.