Gradient-Based Spoof Detection for Biometric Authentication

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Biometric authentication systems face challenges in distinguishing between genuine and spoofed fingerprint images, particularly due to variations in fingerprint patterns and the similarity of spoofed images to real fingers, which can lead to unauthorized access.

Innovation Solution

A gradient-based metric for spoof detection is introduced, utilizing a histogram of variance of gradients computed from input images, which helps differentiate between live and spoofed fingerprints by analyzing minute changes in gradient direction, and is applied to fingerprint ridges to enhance detection accuracy and independence from image size.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If traditional fingerprint authentication is used, then authentication speed is maintained, but spoof detection accuracy deteriorates due to inability to distinguish replica fingerprints

Engineering Contradiction:
Improvespoof detection accuracyVSAvoidauthentication system complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent divides the fingerprint image into multiple blocks and computes gradient variance for each block individually. This segmentation allows the system to analyze local variations in gradient directions across different regions of the fingerprint, enabling detection of spoofing artifacts without requiring a complete system redesign.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent transitions from analyzing traditional fingerprint features to computing gradient images and their variances. By transforming the fingerprint data into gradient space and analyzing variance distributions, the system adds a new dimensional approach to spoof detection that enhances accuracy without increasing physical device complexity.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Measurement precision

If gradient-based metric with histogram of variance is computed, then spoof detection accuracy is improved, but computational complexity increases

Engineering Contradiction:
Improvespoof detection precisionVSAvoidcomputational complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the gradient image into blocks and computes variance for each block, then creates a histogram of these variance values. This segmented approach allows efficient computation by processing smaller regions independently, reducing overall computational complexity while maintaining high measurement precision through aggregated statistical analysis.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent replaces complex mechanical or hardware-based spoof detection mechanisms with computational image processing. By using gradient calculations and variance analysis on digital images, the system achieves high detection precision through software-based methods rather than requiring complex physical verification systems.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

3Reliability

If fingerprint authentication system is made more secure against spoofs, then reliability improves, but ease of operation deteriorates due to additional verification steps

Engineering Contradiction:
Improveauthentication securityVSAvoiduser operation simplicity
Core Design Contradiction:
ReliabilityVSEase of operation

Solution Approach 1:

The patent implements spoof detection that operates automatically as part of the standard authentication flow. The gradient-based metric and histogram analysis are computed without requiring user awareness or additional actions, making the security enhancement transparent and maintaining ease of operation while improving reliability.

Inventive Principle:
Principle #25Self-service

Data Source

PatentUS10176362B1Systems and methods for a gradient-based metric for spoof detection
Publication Date: 2019.01.08 SYNAPTICS INC
  • US10176362B1 patent drawing
  • US10176362B1 patent drawing
  • US10176362B1 patent drawing

AI summary

A system and method for performing spoof detection are disclosed. The method includes: receiving an input image of a biometric; computing a gradient image of the biometric based on the input image of the biometric; dividing the gradient image into blocks of pixels; computing, for each block, a variance value corresponding to a variance of the pixels in the block; and, determining whether the input image is a replica of the biometric based on the variance values of the blocks of the gradient image.