"Healthy surveillance": Designing a concept for privacy-preserving mask recognition AI in the age of pandemics

Abstract

The obligation to wear masks in times of pandemics reduces the risk of spreading viruses. In case of the COVID-19 pandemic in 2020, many governments recommended or even obligated their citizens to wear masks as an effective countermeasure.In order to continuously monitor the compliance of this policy measure in public spaces like restaurants or tram stations by public authorities, one scalable and automatable option depicts the application of surveillance systems, i.e., CCTV. However, large-scale monitoring of mask recognition does not only require a well-performing Artificial Intelligence, but also ensure that no privacy issues are introduced, as surveillance is a deterrent for citizens and regulations like General Data Protection Regulation (GDPR) demand strict regulations of such personal data. In this work, we show how a privacy-preserving mask recognition artifact could look like, demonstrate different options for implementation and evaluate performances. Our conceptual deep-learning based Artificial Intelligence is able to achieve detection performances between 95% and 99% in a privacy-friendly setting. On that basis, we elaborate on the trade-off between the level of privacy preservation and Artificial Intelligence performance, i.e. the "price of privacy".

Face Mask Detection in Smart Cities Using Deep and Transfer Learning: Lessons Learned from the COVID-19 Pandemic

“…Large-scale monitoring and FMD of individuals in public areas do not only need well-performing AI tools but also require to include privacy preservation modules. This is because any kind of surveillance can violate citizens and regulations, including the general data protection regulation (GDPR), which sets strict regulations on the use and share of personal [11]. Although FMD developers deny privacy problems that come with FMD systems as they do not identify individuals in most cases, and detect the existence of face masks, adding privacy protection mechanisms to them will increase the users' trust and acceptance of the FMD technology.…”

“…To cope with this challenge, the authors in [11], explain a privacy-protection FMD system that might be developed, demonstrating various implementation and performance evaluation options.…”

“…In addition to the above, maintaining people's privacy during the mask detection task is another critical issue [10]. To that end, various studies have been proposed to develop face detection systems using AI-based video analytics, ensuring that only the proper wearing of masks is detected and individual privacy is not exposed [11]. The success of artificial intelligence (AI) and machine learning (ML) models in object detection and face recognition makes this technology suitable for developing face mask detection techniques [12].…”