Audio-Visual PIN Entry Prevents OCR Capture
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for entering personal identification codes in graphical user interfaces, such as virtual Pin Pads, are vulnerable to attacks using simple graphical capture and optical recognition techniques, which can compromise security by allowing unauthorized access to the entered codes.
Innovation Solution
A method that combines a graphical user interface with an independent audio channel to securely enter personal identification codes by displaying reference elements and characters, where the audio channel indicates the correct reference element for each character, making it difficult for attackers to use OCR techniques, and optionally includes randomization and timeout validation to enhance security.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a virtual Pin Pad is used for secure entry, then the personal identification code can be securely entered and encrypted, but the interface becomes vulnerable to graphical capture and optical recognition attacks
Solution Approach 1:
The patent replaces the purely graphical visual channel with a multi-sensory interface that includes audio feedback. The system substitutes static graphical elements with dynamic audio-visual correspondence, where reference elements are announced through audio channels and must be matched with visual characters on the screen, making simple graphical capture and OCR attacks ineffective
Solution Approach 2:
The patent adds an audio dimension to the traditionally two-dimensional graphical interface. By introducing audio announcements of reference elements and requiring users to match them with visual characters, the system creates a multi-dimensional interaction that cannot be captured by simple graphical screenshots or OCR techniques
2Ease of operation
If reference elements and characters are displayed on the graphical interface, then the user can visually match them, but attackers can capture and analyze these zones using OCR techniques
Solution Approach 1:
The patent introduces audio announcements as an intermediary between the reference elements and the user. Instead of directly displaying reference elements for visual matching, the system announces them through audio channels, requiring users to match the audio announcements with visual characters on the screen. This intermediary audio layer prevents direct visual capture and OCR analysis of the reference elements
Solution Approach 2:
The patent makes the interface dynamic by requiring real-time audio-visual matching rather than static visual comparison. The audio announcements of reference elements occur in sequence, and users must dynamically match them with the corresponding visual characters before time expires, creating a time-sensitive interactive process that cannot be captured by static graphical screenshots
3Reliability
If audio channel is added to indicate reference elements, then security against OCR attacks is improved, but device complexity increases
Solution Approach 1:
The patent utilizes the audio channel of the mobile device, which is already a standard component for notifications and alerts, to announce reference elements. By repurposing the existing audio capability for security enhancement rather than adding a completely new input modality, the system achieves improved security without significantly increasing device complexity
Solution Approach 2:
The system uses the device's own audio output to announce reference elements, making the device self-sufficient for the enhanced security function. The mobile device's built-in speaker and audio processing capabilities are leveraged to provide audio feedback, eliminating the need for external audio equipment or additional hardware components
4Object-affected harmful factors
If randomization of reference elements and characters is implemented, then attack complexity increases, but ease of operation decreases
Solution Approach 1:
The patent implements randomization of reference elements and characters that is reset with each new personal identification code entry. This periodic re-randomization ensures that even if an attacker captures a graphical state, it becomes invalid for the next code entry, increasing attack complexity while maintaining ease of operation through the familiar audio-visual matching process
Data Source
Figure 1~3
Figure 4~5
Figure 6
AI summary
The present invention relates to method(M) for entering a personal identification code composed of characters in a device comprising a user graphical interface. Said method is characterized in that it comprises the steps of: -displaying a plurality of reference elements on the device's user graphical interface; -displaying a plurality of characters on the device's user graphical interface; -indicating, by means of an audio canal of the device, a reference element to be used for setting a personal identification code's character in correspondence with said reference element; and -capturing said personal identification code's character which has been positioned by means of said user graphical interface to correspond with said reference element.