Dynamic Vision Sensor Face Authentication
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional facial authentication technologies face challenges in distinguishing between real human faces and images or models, leading to potential unauthorized access.
Innovation Solution
An electronic device equipped with an image sensor and a dynamic vision sensor (DVS) that detects radial movements in a user's face, such as changes in iris size, to verify the authenticity of the face, thereby differentiating between real and photographic or modeled faces.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional facial authentication technology is used, then facial authentication can be performed, but it cannot distinguish between real human faces and images or models
Solution Approach 1:
A dynamic vision sensor (DVS) is introduced as an intermediary device between the image sensor and the authentication system. The DVS captures dynamic movement information that serves as a mediator to distinguish real faces from static images or models, enabling the system to detect temporal changes in facial features that conventional image sensors cannot capture
Solution Approach 2:
The system transitions from static facial recognition using conventional image sensors to dynamic facial analysis using DVS. By capturing motion vectors and temporal variations in facial movements, the system can detect whether the input is a real living face or a static image/model, significantly improving authentication reliability
2Reliability
If additional sensors are added to improve authentication accuracy, then security is enhanced, but device complexity increases
Solution Approach 1:
The dynamic vision sensor serves multiple functions: it detects facial movements for authentication, captures dynamic characteristics for liveness detection, and provides motion vector data for behavioral analysis. This multi-functionality reduces the need for multiple separate sensors, thereby enhancing security while limiting the increase in device complexity
Solution Approach 2:
The patent combines the DVS with the existing image sensor system into an integrated facial recognition module. By merging the dynamic movement detection capability with the static image capture function, the system achieves enhanced authentication security without requiring completely separate sensor systems, thus managing device complexity
3Speed
If dynamic vision sensor data is processed, then real-time authentication is enabled, but data processing volume increases
Solution Approach 1:
The system extracts only the essential dynamic movement information from the DVS data, such as motion vectors, movement direction, and temporal changes, rather than processing the complete raw data stream. This extraction approach enables real-time authentication by focusing computation on the most critical features that indicate liveness
Solution Approach 2:
The system processes only the necessary portion of DVS data required for authentication decisions, such as movement detection in specific facial regions and temporal pattern analysis, rather than processing all captured data. This partial processing strategy maintains fast authentication speed while reducing the effective data volume that requires processing
Data Source
AI summary
An example electronic device may receive user input in a state in which security of the electronic device is configured; based on receiving of the user input, display a designated screen on a display; acquire image data including a face of a user through an image sensor; detect a radial movement from a part of the face using a DVS while acquiring the image data; detect first face data corresponding to the face from the image data and compare the detected first face data with second face data stored in the electronic device; and, based on the detection of the radial movement and correspondence between the first face data and the second face data, release the state in which the security is configured.