Many contagious diseases such as Covid19 can induce high temperatures and fever in a significant number of affected individuals. Contact tracing of such individuals and analysis of their behaviour and interactions with others and their environment can be a useful tool to contain spread of contagious diseases. This is in particular useful for back-to-work strategy as well as protection of critical personnel or more vulnerable individuals. Existing solutions are either too expensive (e.g. high-end thermal cameras) or not precise enough (e.g. contact tracing using smartphones). Also, surveillance and analysis of their content pose various ethical challenges, including invasion of privacy.
What it does
We propose a solution that allows for efficient tracking of individuals potentially infected by a transmissible disease while their identity is protected at the same time. However, it will be possible to recover the identity using cryptographic keys if a consent is given or when an authority makes a request. A key component of the system is an innovative camera that combines infrared/thermal and visible light sensors while securing the visible light images. The camera sends captured images using real-time streaming to a server while it protects the anonymity of individuals in the scene by hiding the visible light images inside thermal images using a cryptographic tool called transmorphing. The proposed solution is more accurate when compared to solutions which use smartphone positions or relative distance between devices through Bluetooth discovery, to trace proximity between individuals. It also does not require that individuals carry a smartphone or install a specific app. The camera is low-cost when compared to existing thermal camera solutions to measure temperature from distance.
How we built it
We designed and built a new type of connected camera with multiple sensors which can track people while protecting their privacy and at the same time identify those with high body temperature. The device is in form of a kitset made of off-shelf components and open software that will allow anybody to build it at low cost, easily install, and start using it. A dedicated server records all captured footage in a secure and anonymized way with possibility of further analysis, visualization and eventual de-anonymization. We also created an innovative custom camera enclosure to protect its components while making it easy to install anywhere.
Challenges we ran into
As low-cost infrared/thermal sensors have lower resolution and precision, we had to combine three such sensors in order to be able to increase both quality and precisions by relying on advanced image processing algorithms.
Accomplishments that we're proud of
- A complete design of hardware architecture of the camera from off-shelf components.
- Visible light as well as infrared/thermal video capture and streaming based on existing open source software.
- Enhancement of the software by including a revolutionary reversible anonymization solution made in EPFL!
- A customized design of the camera enclosure enabling its easy installation anywhere.
What we learned
A multidisciplinary team of committed and hardworking members, enhanced by efficient remote working solutions and a clear common objective, can come up with an innovative camera from scratch and have it ready to build in 72 hours!!!
Three EPFL students will continue the work during their semester projects in order to build several cameras and install them at selected locations for trials. We plan to also further enhance the system to perform AI-based advanced video analytics, to trace people between cameras and to merge with other contact tracing approaches such as those based on smartphones. This hackathon is the first phase of a more ambitious project in three phases.