User Authentication via CNN Drawing Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing authentication methods, particularly those relying on passwords and PINs, are inconvenient for users and susceptible to attacks such as social engineering and shoulder surfing, leading to security concerns and user experience issues.
Innovation Solution
A method that combines static image pattern recognition with dynamic behavioral analysis using convolutional neural networks (CNN) to identify and authenticate users based on unconstrained free-draw procedures on interactive surfaces, providing a resistant solution to simple reproduction attacks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If traditional authentication methods (passwords, PINs) are used, then ease of operation is improved, but security and susceptibility to attacks worsen
Solution Approach 1:
The patent replaces traditional mechanical/password-based authentication with biometric authentication using convolutional neural networks to analyze drawings. The system captures drawing characteristics (stroke order, pressure, speed) and uses CNNs to verify user identity, eliminating the need for passwords while improving both security and ease of use.
Solution Approach 2:
The patent changes the authentication parameters from discrete values (passwords, PINs) to continuous biometric parameters (drawing trajectory, stroke dynamics, pressure variations). This allows for more secure and flexible authentication by analyzing the unique way users perform drawing actions rather than relying on memorized codes.
2Reliability
If biometric authentication is used, then security is improved, but device complexity and processing requirements worsen
Solution Approach 1:
The patent segments the authentication process into distinct modules: drawing capture, feature extraction, and classification. The Convolutional Neural Network is divided into convolutional layers for feature extraction and fully connected layers for classification, allowing each component to be optimized independently and reducing overall system complexity.
Solution Approach 2:
The patent introduces an intermediary feature extraction layer that transforms raw drawing data into meaningful features (stroke order, pressure variations, speed). This intermediary representation simplifies the authentication decision process by providing a condensed summary of drawing characteristics that can be easily compared against stored templates.
3Productivity
If static image analysis is used, then processing speed is improved, but ability to detect behavioral patterns worsens
Solution Approach 1:
The patent transitions from static image analysis to dynamic drawing analysis by capturing temporal information (stroke order, speed, pressure changes over time). The system analyzes the dynamic characteristics of how users perform drawings rather than just the final image, enabling detection of behavioral patterns that are invisible to static analysis.
Solution Approach 2:
The patent performs preliminary feature extraction during the drawing capture phase, continuously analyzing stroke dynamics and extracting behavioral features in real-time. This preliminary action allows the system to prepare authentication decisions before the drawing is complete, improving processing speed while capturing comprehensive behavioral patterns.
Data Source
AI summary
A method of identifying a user comprising: acquiring a signal representing an identification drawing input generated by an unconstrained movement of the user on an interactive surface; generating an image that corresponds to the signal; processing the signal to generate a user behavior vector; and identifying the user by comparing the image with an identification image representing an identification drawing previously stored in association with the user; and comparing the user behavior vector with a previously user behavior vector associated with the identification drawing. Moreover, the present invention also refers to an analogous system and non-transitory computer readable medium for performing the proposed identifying method of the user by an unconstrained free draw on an interactive surface.