The origins go back to the year 2000, when Prof. Drahanský was working on his master's thesis on fingerprint recognition and Dr. Orság on his master's thesis on speech recognition. With the establishment of the STRaDe research group and the gradual increase in the number of students not only at the Ph.D. level, but also at the master's and bachelor's level, we started to work on a rather large group of biometric topics. In the beginning it was fingerprints and voice. Later we left the speech topic, but we started research in the field of liveness detection (mainly on fingers), 3D face recognition, 3D hand and finger or hand veins, production of various fake biometric characteristics, iris and retina sensing and recognition. We are also interested in the effect of disease on the recognition process of biometric features, especially for finger and eye features. Our research includes not only the design and implementation of appropriate algorithms, but also the acquisition of biometric data and the design and construction of hardware for biometric systems. By participating in many biometric projects, including testing of biometric systems (e.g. for the National Security Agency), we have gained a lot of experience that we have gradually shared within the team. We also have experience in commenting on draft working standards.
We also looked at biomedical data. The beginnings of work on biomedical applications date back to 2012, when our STRaDe research group participated in the ICRC (International Clinical Research Center) project on the BME platform, which was part of St. Anne's University Hospital. This project ended in 2015. In the past, we have also worked on applications for processing and analyzing echocardiographic medical records, hematoma detection and sizing for smart devices on the Android platform, and applications for processing the AHEAD database (containing detailed data of patients suffering from some cardiovascular diseases). However, the most interesting topics are the detection and recognition of skin diseases on the fingers and diseases of the retina in particular, but also of the iris of the eye.
Finally, we must not forget another area in which we have been active, namely computer vision and robotics. Computer vision is closely related to biometric systems. To obtain an image of many biometric characteristics, a camera system must be used. The principles of processing the acquired data are often very similar to computer vision algorithms and techniques.
Our greatest successes were achieved within the projects of the Ministry of Industry and Trade of the Czech Republic (research and development of technologies for intelligent optical surveillance systems) and the Ministry of the Interior of the Czech Republic (tools and methods for video and image processing in the fight against terrorism), which resulted in a number of products. The first is a camera system with visibility up to 8 km, mounted on a military manipulator and connected to a powerful computer with GPU. It includes a wide range of applications (algorithms) for e.g. object tracking (including partial movement behind barriers), image stabilization in video, detection and recognition of defined objects (e.g. weapons, faces), anonymization of selected objects in video, 3D data processing, etc.
In the field of computer vision hardware development, we have experience in the design of camera systems for remote object tracking, as well as in the development of printed circuit boards with DSP and FPGAs for PAL/NTSC and FullHD systems. You could say that we have been in this field since the transition from analog to digital cameras.
In 2023 and 2024, the group underwent personnel changes. This led to a reorganization to maintain research in biometrics and to support it with talented students from our faculty.