Structured Light Face Recognition System
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional face recognition systems are hindered by changes in ambient lighting and facial attitude, leading to reduced accuracy in recognition.
Innovation Solution
A system that emits structured light to a face, capturing light-source-illuminated images and extracting features such as 3D coordinates, surface normals, and light absorption characteristics, which are then compared to a face template to improve recognition accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a passive image capture method is used without active light emission, then the device complexity is reduced, but the measurement precision deteriorates due to influence from ambient lighting changes and facial attitude changes
Solution Approach 1:
The system performs preliminary action by actively illuminating the face with structured light patterns before capturing images. Multiple light patterns are projected onto the face in advance, and the corresponding images are captured while the face is in various attitudes. This preliminary capture of light-illuminated images across different orientations enables the system to pre-establish a comprehensive set of facial features that are less sensitive to subsequent lighting and attitude variations during recognition.
Solution Approach 2:
The system transitions from two-dimensional passive image capture to three-dimensional structured light illumination. By projecting structured light patterns and capturing the reflected light from multiple angles, the system obtains three-dimensional facial information including depth and surface geometry. This dimensional enhancement allows the extraction of robust facial features that remain consistent across different facial attitudes and ambient lighting conditions.
2Measurement precision
If structured light illumination is used to improve measurement precision, then the measurement precision improves, but the device complexity increases due to additional light source generation and control mechanisms
Solution Approach 1:
The system employs periodic action by sequentially projecting multiple structured light patterns onto the face at different time intervals. Each light pattern is emitted for a brief period, followed by image capture, then the next pattern is projected. This periodic illumination approach allows the system to gather comprehensive three-dimensional facial information through time-multiplexed measurements, achieving high measurement precision while managing device complexity through temporal sequencing rather than simultaneous multi-source illumination.
3Measurement precision
If multiple light patterns are projected successively to capture comprehensive facial features, then the measurement precision improves, but the loss of time increases due to sequential illumination and capture cycles
Solution Approach 1:
The system maintains continuity of useful action by seamlessly transitioning between projecting different structured light patterns and capturing corresponding images. The sequential illumination and capture operations are performed in rapid succession without interruption, forming a continuous measurement process. This continuous operation minimizes idle time between measurements and ensures that the entire face is captured in various attitudes within a single coherent measurement sequence, reducing overall time loss while maintaining high measurement precision.
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
The system enhances face recognition accuracy by minimizing the impact of ambient lighting and facial orientation changes, resulting in improved consistency with face templates.
Implementation Method 1
a light source generation module operative to emit at least one group of structured light to a face to be recognized
Implementation Method 2
material information indicating a light absorption characteristic of the detection point
Data Source
AI summary
There is disclosed an apparatus, system, method and computer program product for recognizing a face, the method comprising: emitting at least one group of structured light to a face to be recognized, successively; capturing a set of light-source-illuminated images of the face when the face is illuminated successively by each group of light of the at least one group of structured light; extracting a first set of features including a feature of each detection point in a set of detection points of the face based on the set of light-source-illuminated images; acquiring a second set of features including a feature of each detected point in a set of detected points of a face template; computing a similarity between the face and the face template based on the first set of features and the second set of features; and recognizing the face as being consistent with the face template if the similarity is larger than a threshold.