Polarization Filter for Face Recognition in Ambient Light
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Face recognition systems are prone to errors in ambient light environments, particularly when exposed to sunlight, as they struggle to achieve homogeneous illumination, leading to reduced performance and increased detection errors, especially in modern, light-flooded architectures like airports.
Innovation Solution
The use of polarized light filtering, where the system emits and receives light only in a specific polarization plane, using polarizing filters on transparent surfaces and within the imaging system, to suppress stray light and ensure consistent illumination without obstructing the environment with opaque barriers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Illumination intensity
If ambient light is blocked out using opaque barriers or windowless rooms, then homogeneous illumination of the face is achieved, but the environment becomes restricted and modern light-flooded architecture cannot be used
Solution Approach 1:
The patent changes the parameter of light polarization state to differentiate between useful illumination light and disturbing ambient light. By using polarized illumination and corresponding polarizing filters in the imaging path, the system can selectively transmit or block light based on its polarization state, enabling homogeneous face illumination while maintaining environmental openness through transparent surfaces with polarization control
Solution Approach 2:
The patent introduces polarizing filters as intermediary elements between the light sources and the imaging system. These filters act as mediators that selectively transmit polarized illumination light while blocking non-polarized or differently polarized ambient light, thereby achieving homogeneous illumination without requiring opaque barriers
2Illumination intensity
If large lighting is installed to illuminate the face, then sufficient brightness is achieved, but the system becomes more complex and energy-consuming
Solution Approach 1:
The patent applies polarization state as a distinguishing parameter for the illumination light. By using polarized light sources and corresponding polarizing filters in the imaging path, the system achieves sufficient face brightness without requiring large, complex lighting installations, as the polarization filtering enhances the effectiveness of the illumination
3Illumination intensity
If polarizing filters are used to block ambient light, then homogeneous illumination is improved, but transmission losses occur requiring more intensive lighting
Solution Approach 1:
The patent uses polarization state as a selective parameter to differentiate between useful illumination light and disturbing ambient light. By matching the polarization state of the illumination source with the orientation of polarizing filters in the imaging path, the system maximizes the transmission of useful light while blocking ambient light, thereby reducing the energy required for illumination compared to non-polarized systems
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 allows for efficient and accurate person recognition in light-flooded environments by isolating the useful light plane, reducing detection errors, and maintaining high-quality image capture without dazzling the subject or restricting the environment.
Implementation Method 1
a polarization filter element (107), which is arranged at a distance from the illumination device (103) and from the image recording device (105) and is provided to suppress the stray light (102) in the predetermined plane of polarization
Implementation Method 2
an illumination device (103) which is designed to emit polarized illumination light (104) in a predetermined light polarization plane
Implementation Method 3
an image recording device (105) which is designed to receive light (106) in the predetermined light polarization plane
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The present disclosure relates to an image acquisition system (100, 200) for capturing an image of a person (101) for person recognition in an environment with ambient light (102), comprising: an illumination device (103) configured to emit polarized illumination light (104) in a predetermined plane of light polarization (301); an image acquisition device (105) configured to receive light (106) in the predetermined plane of light polarization (301) in order to obtain an image of the person (101); and a polarization filter element (107) arranged spaced apart from the illumination device (103) and the image acquisition device (105) and designed to suppress the ambient light (102) in the predetermined plane of polarization (301).