Privacy-Preserving Facemask Compliance Measurement with Depth Wavelets
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems for measuring facemask compliance in public spaces often rely on visible-light images, which compromise privacy and can be inaccurate, especially for diverse populations. These systems also tend to be computationally expensive and power-intensive, limiting their applicability.
Innovation Solution
The implementation of a privacy-preserving facemask-compliance measurement system that uses depth images captured by devices such as depth cameras or LiDAR. This system generates facial depth images, extracts wavelet descriptors, and analyzes spectral-density values using a classification model to determine mask-wearing compliance without identifying individuals.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If visible-light images are used for facemask compliance measurement, then the system can identify individuals and provide detailed facial information, but privacy is compromised and measurement accuracy decreases for diverse populations
Solution Approach 1:
The patent extracts only the necessary depth information from the full facial structure, focusing solely on the mask-wearing status detection while discarding all identifying facial features. The system processes only the depth data required to determine mask compliance, eliminating the need to capture or store identifiable facial images.
Solution Approach 2:
The patent introduces depth images as an intermediary medium between the camera and the analysis system. Instead of directly analyzing visible-light facial images, the system uses depth images as a intermediary that preserves only the geometric information needed for mask detection while stripping away all privacy-sensitive visual information.
2Measurement precision
If visible-light image processing is used for facemask detection, then detailed facial analysis is possible, but computational cost and power consumption increase
Solution Approach 1:
The patent extracts and processes only the depth channel information from the image data, eliminating the need to process the full-color visible-light image data. This selective processing reduces the computational workload by approximately half while maintaining sufficient accuracy for mask compliance measurement.
Solution Approach 2:
The patent segments the image processing task by separating depth information extraction from color information processing. The system performs processing only on the depth component, which contains the necessary geometric information for mask detection, thereby reducing overall computational and energy requirements.
3Reliability
If comprehensive facial analysis is performed to ensure accurate mask compliance measurement, then measurement reliability improves, but privacy protection decreases
Solution Approach 1:
The patent extracts only the specific depth information necessary for mask compliance determination, avoiding extraction of any identifying facial features. The system processes solely the geometric data related to mask positioning and coverage, completely excluding personally identifiable information from the analysis.
Solution Approach 2:
The patent applies local quality analysis by focusing depth processing only on the specific region where the mask would be worn (the face area), while ignoring all other facial features and characteristics. This localized approach maintains measurement reliability for mask detection while protecting privacy by analyzing only the minimal necessary region.
Data Source
AI summary
Disclosed herein are systems and methods for privacy-preserving facemask-compliance-level measurement. In an embodiment, a mask-compliance measurement system includes a processor that is configured to generate, from an image of a person, a facial depth image of a region of a face of the person, and to generate facial wavelet descriptors from the facial depth image. The processor is also configured to determine spectral-density values of the wavelet descriptors, and to analyze the spectral-density values to generate a mask-wearing-compliance result for the person. In an embodiment, the analyzing includes using a classification model that is trained to classify sets of spectral-density values with respect to facemask wearing in images from which the spectral-density values were derived.