Areas of expertise
Person recognition
Our research in biometrics focuses on developing solutions for recognising individuals based on biometric features, such as fingerprints, facial recognition, finger or hand veins, and more. We have extensive experience in this field.
Liveness detection
The aim is to determine whether a biometric feature belongs to a real person or if it is fake. Our focus is on testing and developing technologies to authenticate biometric features, specifically in the areas of 2D and 3D face and fingerprint representation.
Biometric sensors
Our research group confidently designs and develops algorithms for commercial biometric sensors. We assertively assess client requirements and create a development plan accordingly.
Artificial Intelligence
Following the trends in image processing, where AI has been used extensively recently, we are looking at other applications using AI for camera data processing in addition to the use of neural networks for biometrics.
Current Research
AI accelerated 2D facial recognition
We focus on developing solutions for fast face recognition while maintaining maximum accuracy. We are also working on transforming neural networks for use in edge computing.
3D bloodstream sensing sensor
Our research explores how to utilize the characteristics of finger veins. We have developed a new sensing device that can effectively sense finger veins from multiple angles. Image processing algorithms are being developed.
Hand liveness detection
Together with deepfake detectors, liveness detection is a key component of all modern biometric sensors. In this research, we focus on increasing the robustness against spoofing.
Generation of synthetic biometric data
Synthetic data can be used to train and evaluate machine learning-based algorithms. Our primary focus is on the generation of synthetic fingerprints and facial images.
Multimodal people detection in surveillance video
In practice, there are often situations where it is not possible to detect and recognize a person's face. In this case, it is necessary to focus on other features of the person. The results of this research can be used in security services as well as in commercial applications.
About
The STRaDe research group is part of the Institute of Intelligent Systems at the Faculty of Information Technology of the Brno University of Technology. The main focus of the group is research and development of technologies in the field of security, which includes both software and hardware. Security technologies cover a wide range of applications and devices. Most of our research is focused on biometrics. We are involved in many aspects of biometrics and our knowledge and focus covers a wide range of algorithms and solutions such as fingerprinting and recognition, iris and retina, 3D face and hand, vein recognition, liveness detection of various biometrics, etc. We also focus on developing solutions using image processing and artificial intelligence. We have experience with classical deep neural networks, GANs, LSTM networks and their acceleration.