Glance Sensor Positioning for Mobile Terminal Face Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing face recognition technologies in mobile terminals face challenges such as reduced recognition rates with varying skin colors and lighting conditions, vulnerability to spurious photographs due to the need for specific user actions, and high power consumption and noise interference in voice recognition methods.
Innovation Solution
A method for setting an optimal disposition of a glance sensor in a mobile terminal, using metadata from both a front camera and the glance sensor to enhance face recognition security and convenience, where the glance sensor is tilted and positioned adjacent to the front camera to maximize the overlap area for efficient user authentication.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If face recognition is performed using skin color of face, then user authentication can be performed in non-contact manner, but recognition rate is reduced when various skin colors and lighting conditions change
Solution Approach 1:
The patent transitions from 2D skin color analysis to 3D facial structure analysis by introducing a glance sensor that captures depth information. This dimensional change enables recognition based on geometric features rather than color, resolving the contradiction between non-contact operation and recognition accuracy under varying conditions.
Solution Approach 2:
The patent changes the recognition parameter from skin color (2D optical property) to facial depth structure (3D geometric property). By using a glance sensor to capture time-of-flight data, the system switches to parameters that are invariant to lighting and skin color variations, maintaining both non-contact operation and high recognition rates.
2Reliability
If specific user action such as winking or smiling is induced for face recognition, then security against spurious photograph is high, but convenience is reduced and security is vulnerable since action can be recorded
Solution Approach 1:
The system performs preliminary 3D facial structure capture and analysis before authentication is completed. By pre-establishing the user's facial geometry model and comparing it in real-time, the system provides continuous security verification without requiring additional user actions, thus maintaining both security and convenience.
Solution Approach 2:
The patent introduces 3D facial structure data as an intermediary between the user and the authentication system. This intermediary provides continuous verification through depth mapping, eliminating the need for discrete action-based challenges while maintaining high security against photograph attacks.
3Measurement precision
If camera and microphone are used for face and voice recognition, then recognition can be performed, but power consumption is high and voice recognition rate is reduced by external noise
Solution Approach 1:
The patent extracts and utilizes the infrared illumination source and time-of-flight sensing capability from the glance sensor, separating this function from traditional camera-microphone-based recognition. This extraction enables low-power 3D facial recognition that is immune to external noise, reducing both power consumption and noise vulnerability.
Solution Approach 2:
The patent replaces the acoustic field-based voice recognition (microphone) with optical field-based time-of-flight measurement (glance sensor). This substitution eliminates susceptibility to external noise and reduces power consumption by using infrared LED illumination combined with TOF sensing rather than continuous audio processing.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach reinforces security and convenience by allowing user authentication through gazing at the front camera without additional actions, enabling quicker and more accurate face recognition.
Implementation Method 1
a glance sensor tilted by a certain angle and disposed adjacent to the front camera to obtain metadata of the 2D face image
Data Source
Figure 1A
Figure 1B~1C
Figure 2~3
AI summary
Disclosed is a mobile terminal. The mobile terminal may include a front camera obtaining a 2D face image of a user, a glance sensor tilted by a certain angle and disposed adjacent to the front camera to obtain metadata of the 2D face image, and a controller obtaining a distance between the glance sensor and the front camera, the distance enabling an area of an overlap region, where a first region representing a range photographable by the front camera overlaps a second region representing a range photographable by the glance sensor, to be the maximum.