Research in this field focuses on designing methods and algorithms of intelligent processing multidimensional and multimodal data. It encompasses the processing of data from the environment, traffic, autonomous vehicles, speech, image, bioelectrical, biomedical, biometric signals, data in telecommunications and radio-communication networks; machine, device, and material diagnostics and tele-diagnostics, including telemedicine; multimedia and data processing in IoT systems and Smart City, as well as in military techniques, forensics, and cultural heritage. Particular applications involve intelligent processing of audio and video streams, including feature extraction and classification, as well as recommendation systems. The field also involves processing of multidimensional (3D and 4D) spatial data. The term spatial data refers to geographic data of nautical, land, and aerial origin, which is primarily processed with the use of Geographic Information Systems (GIS). To this end, research topics include applications of GIS and spatial analysis in the protection of critical infrastructure, specialized applications of navigational equipment and systems (e.g., for analysis of ship sailing patterns or supporting the visually impaired in their daily routines), and processing and analysis of remote sensing data such as satellite imagery (e.g., for flood detection and analysis), LiDAR (e.g., for 3D shape recognition and reconstruction) and underwater acoustics data (e.g., for detection and classification of seabed types based on single and multibeam sonar echo characteristics). The latter also involves the design of advanced acoustic survey methods for monitoring and mapping underwater resources (such as flora and fauna), pollution and other elements of the marine environment.
From the computer science perspective, research in this field focuses on design of task-efficient software architectures and implementing new and more effective data processing and analysis algorithms. Moreover it also involves the application of machine learning (including baseline and deep learning, semi-supervised learning, transfer learning, learning with weak labels, learning interpretable models, etc.), correlation, and advanced statistical modeling methods.
Keywords:
multidimensional and multimodal data processing, machine learning, advanced statistical learning/modeling methods, data in telecommunications and radio-communication networks; GIS, remote sensing, LiDAR
