Gradient Residual Analysis for Spoof Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Biometric authentication systems face challenges in distinguishing between live and spoofed fingerprint images, as spoof methods can produce images with irregular gradient distributions that resemble real fingers, potentially leading to unauthorized access.
Innovation Solution
A method involving the application of two different convolution kernels to an input image, generating a gradient residual image, and applying a density estimation procedure to identify areas of varied density, determining whether the image is a replica of a biometric based on the results, using a classifier to produce a liveness score.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional fingerprint authentication is used, then authentication speed is fast, but security is compromised due to spoofing attacks
Solution Approach 1:
The patent segments the gradient analysis into multiple directional components (horizontal, vertical, diagonal gradients) and processes them separately through convolution operations. This segmentation allows the system to capture different aspects of gradient distribution independently, improving spoof detection accuracy while maintaining computational efficiency through structured processing.
Solution Approach 2:
The patent transforms the 2D fingerprint image analysis into a 3D feature space by computing gradient distributions across multiple directions and scales. This dimensional transformation enables the system to detect spoofing artifacts that are invisible in the original 2D image space, enhancing security without requiring complex hardware modifications.
2Measurement precision
If gradient distribution analysis is applied to detect spoofs, then spoof detection accuracy is improved, but computational complexity increases
Solution Approach 1:
The patent applies convolution operations selectively to specific regions and gradient directions rather than processing the entire image uniformly. By focusing computational resources on the most discriminative gradient components, the system achieves high spoof detection accuracy while reducing overall computational power consumption compared to exhaustive analysis methods.
Data Source
AI summary
A system and method for performing spoof detection are disclosed. The method includes: receiving an input image of a biometric; generating a first filtered image by applying a first convolution to the input image based on a first convolution kernel; generating a second filtered image by applying a second convolution to the input image based on a second convolution kernel; computing a gradient residual image by subtracting, for each pixel location of the first and second filtered images, a pixel value in the second filtered image from a corresponding pixel value in the first filtered image; applying a density estimation procedure to the gradient residual image to identify areas of varied density; and, determining whether the input image is a replica of the biometric based on results of the density estimation procedure.