The main goal of our tracking system is to provide a reliable and optimized environment for organizations. While technologies that businesses deal with are improving, the development of work environments has been postponed and traditional solutions have continued to be used to address problems. During this process, health violations occurred, and physical contact was not minimized. Our project aims to protect places that require security without the need for contact, costly equipment, or personnel, using new generation technologies.
We stayed loyal to the Python programming language in our project, and used many Python libraries. We trained our highly successful models that perform face recognition, which is the backbone of our project, using the deep learning library TensorFlow. During this process, we were able to transfer the images we obtained from the peripheral units to our models that perform successful recognition using the image processing library named OpenCV. We preferred the PyQt5 library, which provides a graphical user interface experience for effective use by users and administrators of the project. For application management, user activities, and data storage, we preferred Firebase technology.
· MaskNet model that detects masks
· Producer model developed with GAN architecture that makes the individual's masked image unmasked
· Model that performs face recognition on the created unmasked images
Project Presentation Link
Project masknet.h5 Link