Spoofing Detection Using Convolutional Neural Networks

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Facial recognition systems are vulnerable to spoofing attacks using printed photos, face cutouts, digital images, and 3D masks, which pose security risks by allowing unauthorized access.

Innovation Solution

A method and apparatus that utilize a trained machine learning process, specifically a convolutional neural network (CNN), to analyze image data and generate intensity values to determine if an image depicts an unauthentic object, thereby identifying and preventing unauthorized access.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If traditional facial recognition systems are used, then authentication can be performed, but the system becomes vulnerable to spoofing attacks

Engineering Contradiction:

Improveauthentication securityVSAvoidspoofing attack vulnerability

Core Design Contradiction:

ReliabilityVSObject-affected harmful factors

Solution Approach 1:

The patent introduces an intermediary detection system that sits between the image capture and authentication processes. This intermediary analyzes image characteristics (lighting conditions, shadows, reflections, depth information) to determine whether the input is a genuine face or a spoof attempt, thereby mediating the authentication decision and preventing spoofing attacks without compromising the original authentication flow

Inventive Principle:

Principle #24Intermediary (Mediator)

2Reliability

If advanced spoofing detection methods are implemented, then security against spoofing attacks improves, but processing power requirements increase

Engineering Contradiction:

Improvespoofing detection accuracyVSAvoidprocessing power consumption

Core Design Contradiction:

ReliabilityVSUse of energy by moving object

Solution Approach 1:

The detection system is segmented into multiple analysis modules, each examining specific characteristics of the input image (lighting, shadows, reflections, depth). This segmentation allows the system to process only the most relevant features using appropriate algorithms, reducing overall computational burden while maintaining comprehensive detection capability

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

Different regions of the input data are analyzed with different levels of computational intensity based on their importance. Critical authentication regions receive more thorough analysis while less critical areas use lighter processing, optimizing the balance between detection accuracy and processing power consumption

Inventive Principle:

Principle #3Local quality

Data Source

PatentUS12142084B2Apparatus and methods for spoofing detection using machine learning processes

Publication Date: 2024.11.12 QUALCOMM INC

AI summary

Methods, systems, and apparatuses are provided to automatically determine whether an image is spoofed. For example, a computing device may obtain an image, and may execute a trained convolutional neural network to ingest elements of the image. Further, and based on the ingested elements of the image, the executed trained convolutional neural network generates an output map that includes a plurality of intensity values. In some examples, the trained convolutional neural network includes a plurality of down sampling layers, a plurality of up sampling layers, and a plurality of joint spatial and channel attention layers. Further, the computing device may determine whether the image is spoofed based on the plurality of intensity values. The computing device may also generate output data based on the determination of whether the image is spoofed, and may store the output data within a data repository.