The best work in the Young Masters Competition was presented by Kacper Podbucki and Jakub Suder. The award in the Substantially Best Work was awarded by the Program Council of the Forum and the Mazovian Division of the Polish Information Technology Society for the project "CCTV based system for detection of anti-virus masks, based on teaching deep neural networks".

As the authors of the project argue, Kacper Podbucki and Jakub Suder, in the project, wanted to offer inexpensive and accessible prevention and surveillance technology, aimed both at the general public and particular communities, such as highly populated places, shopping centres and small businesses.

“The designed system consists of a camera module and a microcomputer, it is equipped with face detection software and includes a trained convolutional neural network model for validating the correct application of masks on the face. The main advantages of our solution are mobility, compact size of the system, high reliability and accuracy ”- says Jakub Suder.

The doctoral student at the Doctoral School of the Poznan University of Technology adds that the project resulted in the design, construction and testing of a system for monitoring and detecting the correct application of an anti-virus mask on the face. Thanks to inexpensive hardware -  a Raspberry Pi 4 model B microcomputer and a camera module, we can create a high-quality and reliable monitoring system for the currently important aspect of preventing the spread of the COVID-19 SARS-CoV-2 pandemic.

The presented system was presented at two international conferences:

• IBM IEEE CASEDS - AI Compute Symposium (Zurich, Switzerland), the presentation distinguished and addressed to a special issue of the Journal on Emerging and Selected Topics in Circuits and Systems
• 24th Conference SPA 2020. Signal Processing: Algorithms, Architectures, Arrangements, and Applications, the presentation can be found in the IEEE Xplore database: “CCTV based system for detection of anti-virus masks” https://doi.org/10.23919/SPA50552 .2020.9241303

The scientific supervisors of the work were prof. Adam Dąbrowski and Tomasz Marciniak, PhD.

Congratulations!

From the left: Tomasz Marciniak, PhD,  Jakub Suder, Kacper Podbucki and prof. Adam Dąbrowski