Artificial Fingerprint Data Encoding for Robust Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing two-dimensional codes, such as QR codes, are vulnerable to damage and tampering, making it difficult to extract information and are susceptible to cyberattacks, while biometric fingerprints face privacy concerns and require specialized devices for data extraction.
Innovation Solution
The development of artificial fingerprints encoded with data using a processor and non-transitory computer-readable medium, which generates a minutiae template from a machine-readable image, allowing for extraction of information using commonly available imaging devices without privacy concerns.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional two-dimensional codes (QR codes, matrix codes) are used for data encoding, then data can be quickly accessed using camera scanners, but the codes are vulnerable to damage and tampering making information extraction difficult
Solution Approach 1:
The patent creates artificial fingerprint images that replicate the visual appearance of real fingerprints but are synthetically generated. These artificial copies contain embedded data in the form of minutiae point characteristics (location, orientation, type) that can be extracted and verified, providing reliable data extraction even when the image is partially damaged or altered.
Solution Approach 2:
The fingerprint data is segmented into multiple minutiae points, each with independent characteristics (x-coordinate, y-coordinate, orientation, type). This segmentation allows the system to extract data from individual minutiae points even when other parts of the fingerprint image are damaged or tampered with, improving overall data extraction reliability.
2Ease of operation
If biometric fingerprints are used for data encoding, then data can be extracted using specialized devices, but privacy concerns arise and specialized devices are required
Solution Approach 1:
The artificial fingerprint images are designed to be readable by standard camera scanners and image processing systems, the same devices used for traditional QR codes and matrix codes. The minutiae point characteristics are encoded in the visual pattern itself, allowing any device capable of capturing and processing images to extract the embedded data without requiring specialized biometric scanning equipment.
Solution Approach 2:
The patent replaces the need for specialized optical sensors and contactless fingerprint scanners with standard camera-based imaging systems. The data is encoded in the visual pattern of the artificial fingerprint, which can be captured by any camera and processed through image analysis algorithms to extract minutiae point characteristics, eliminating the need for dedicated biometric hardware.
3Reliability
If biometric fingerprints are used for data encoding, then unique identification is achieved, but privacy issues and security risks arise
Solution Approach 1:
Instead of using actual biometric fingerprint data from individuals, the system generates artificial fingerprint images that visually resemble fingerprints but contain no real biometric information. The unique identification capability is achieved through synthetically generated minutiae point patterns, eliminating privacy concerns associated with collecting, storing, and processing real biometric data while maintaining the uniqueness and reliability of fingerprint-based identification.
Data Source
AI summary
Systems and methods of extracting information encoded in a machine-readable image (e.g., an artificial fingerprint) are disclosed. The methods include receiving a machine-readable image and generating a minutiae template from the machine-readable image. The minutiae template includes a plurality of minutiae points, each being representative of a local ridge discontinuity in a human fingerprint. The methods further include identifying, for each of the plurality of minutiae points, a minutia point orientation and a minutia point type, and retrieving, from a data store, a plurality of blocks associated with the identified minutiae point orientations and minutiae point types where each block includes two or more bit values. The retrieved plurality of blocks are concatenated to generate a bit stream encoded in the machine-readable image.