Dynamic Signature Recognition Threshold Adjustment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing signature verification methods on electronic devices face challenges in accuracy and convenience due to the difficulty in regularly signing autographs, fixed signature recognition training times, and failure to utilize environmental information associated with signature input.
Innovation Solution
The method involves generating and utilizing signature recognition data by cumulating user signature data with previously stored training data, storing information for distinguishing signature recognition states, and adjusting the signature recognition standard based on the accuracy of additional signature recognition data, while considering environmental information during signature input.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a fixed number of signature recognition training times is used, then the authentication process is simplified, but the recognition accuracy decreases because various features of each user's signature cannot be fully reflected
Solution Approach 1:
The patent implements dynamic adjustment of the training times threshold based on user authentication history. The threshold is not fixed but adapts over time, increasing for users with successful authentication records and decreasing for those with failures, thereby optimizing both accuracy and adaptability without requiring excessively complex fixed thresholds
Solution Approach 2:
The system changes the parameter of training times threshold dynamically based on authentication results. By adjusting this parameter according to user performance and environmental factors, the system achieves high recognition accuracy while maintaining operational simplicity
2Reliability
If signature verification is performed frequently to improve accuracy, then recognition reliability increases, but user convenience decreases due to the difficulty of regularly signing autographs
Solution Approach 1:
The system performs preliminary signature training during device setup or idle periods, accumulating signature data beforehand. This preliminary action reduces the need for frequent verification signatures during actual authentication, improving convenience while maintaining accuracy through pre-collected biometric data
Solution Approach 2:
The system automatically accumulates and processes signature data without requiring active user participation in training exercises. The device self-updates the recognition model using stored data, eliminating the burden of regular user signing while maintaining high verification accuracy
3Device complexity
If environmental information associated with signature input is not utilized, then the authentication process is simpler, but the recognition rate of user signatures decreases
Solution Approach 1:
The system uses the touch screen to serve multiple functions: as the display interface, as the signature input surface, and as the sensing element for capturing signature data and environmental context. This multi-functionality allows environmental information utilization without adding separate hardware complexity
Solution Approach 2:
The touch screen acts as an intermediary that captures both the signature itself and environmental context (pressure, location, timing). This intermediary element translates physical signing actions into digital data that includes environmental features, enabling enhanced recognition without increasing overall system complexity
Data Source
AI summary
An operation method of an electronic device is provided. The operation method includes registering, as reference signature data, at least one handwritten signature inputted into the electronic device by a user input means, authenticating an inputted handwritten signature by comparing data of the inputted handwritten signature with the registered reference signature data when the handwritten signature is inputted by the user input means, and further registering, as reference signature data, handwritten signature data regarding the inputted handwritten signature when the inputted handwritten signature is authenticated normally as a result of the authentication.