Jensen, Ulf

Digital Sports

The Digital Sports Team at the Pattern Recognition Lab deals with the processing and analysis of kinetic, kinematic and physiological data for sports and health research. Our goal is to develop innovative systems in close cooperation with industry and healthcare facilities to support, guide and motivate athletes, patients and seniors. To reach this goal, mathematically funded state-of-the-art methods in signal processing, data mining and pattern recognition are applied and developed.

From an application perspective, our activities can be roughly divided into two parts. On the one hand, we perform offline analysis of large datasets. Pattern recognition and machine learning methods are capable of answering research questions in biomechanics or sport science. To support people during activity, on the other hand, we develop feedback application that perform live data analysis. Our focus are Body Sensor Network (BSN) applications that can be worn during activity.

These live applications have specific technical needs as they run on mobile devices with limited resources. Therefore, our research focusses on methods for embedded classification to adress low computational power and restricted memory while maintaining high discriminative performance.

Applications

The systems we develop have different objectives depending on the application field. In sports and exercise, athletes want to keep a training record, get supported in preventing injuries and optimize performance. Feedback application are capable of tracking and even supporting them while exercising. Regarding the field of healthcare and age, the developed systems can be applied as a diagnosis tool for physicians but also monitor and motivate the users. Activity is a perfect tool to prevent and treat common lifestyle deseases and we develop tool to assess the preparedness of an individual for activity, to track the activity levels for desease monitoring and to motivate individuals via social networking. In research and testing, we support different projects in biomechanical data analysis and develop mobile recording systems. Sports are normally not performed in a research lab and real-life data helps to better understand the needs of an athlete. Therefore, flexible, mobile und sufficiently accurate data recording is needed.

Development Platform

In a lab environment, highly accurate and comprehensive 3-D motion capture systems established as gold standard for movement analysis. However, for real-world data recording and processing, there is a high demand for mobile, lightweight, small and accurate solutions. We employ the capabilities of Body Sensor Networks (BSN) for data recording and live processing to provide research tools and develop innovative mobile applications. The key factors from an application perspective are usability, accuracy and price.

The Pattern Recognition Lab uses BSN consisting of Shimmer Sensor nodes and Android-based mobile devices. The system is highly flexible as it can integrate different kinematic (Accelerometer, Gyroscope) and physiological (ECG, EMG) sensors and allows direct access to the unprocessed sensor data.

Contact

Address

Dr.-Ing. Ulf Jensen

Alumnus of the Pattern Recognition Lab of the Friedrich-Alexander-Universität Erlangen-Nürnberg



	Website deprecated and outdated. Click here for the new site.

Department of Computer Science 5 Our Team Jensen, Ulf Projects Lectures Publications Thesis Curriculum Vitae Research Publications Free Software Data Courses Curriculum Theses Press Releases Cooperations Open Positions LME Videos Ph.D. Gallery Contact Intranet Impressum Datenschutzerklärung Contact Address Universität Erlangen-Nürnberg Chair of Computer Science 5 (Pattern Recognition) Haberstr. 2 91058 Erlangen Germany Driving directions Powered by	Dept. of Computer Sc. » Pattern Recognition » Our Team » Jensen, Ulf Dr.-Ing. Ulf Jensen Alumnus of the Pattern Recognition Lab of the Friedrich-Alexander-Universität Erlangen-Nürnberg The idea is really really simple, but what you get is almost like magic Digital Sports The Digital Sports Team at the Pattern Recognition Lab deals with the processing and analysis of kinetic, kinematic and physiological data for sports and health research. Our goal is to develop innovative systems in close cooperation with industry and healthcare facilities to support, guide and motivate athletes, patients and seniors. To reach this goal, mathematically funded state-of-the-art methods in signal processing, data mining and pattern recognition are applied and developed. From an application perspective, our activities can be roughly divided into two parts. On the one hand, we perform offline analysis of large datasets. Pattern recognition and machine learning methods are capable of answering research questions in biomechanics or sport science. To support people during activity, on the other hand, we develop feedback application that perform live data analysis. Our focus are Body Sensor Network (BSN) applications that can be worn during activity. These live applications have specific technical needs as they run on mobile devices with limited resources. Therefore, our research focusses on methods for embedded classification to adress low computational power and restricted memory while maintaining high discriminative performance. Applications The systems we develop have different objectives depending on the application field. In sports and exercise, athletes want to keep a training record, get supported in preventing injuries and optimize performance. Feedback application are capable of tracking and even supporting them while exercising. Regarding the field of healthcare and age, the developed systems can be applied as a diagnosis tool for physicians but also monitor and motivate the users. Activity is a perfect tool to prevent and treat common lifestyle deseases and we develop tool to assess the preparedness of an individual for activity, to track the activity levels for desease monitoring and to motivate individuals via social networking. In research and testing, we support different projects in biomechanical data analysis and develop mobile recording systems. Sports are normally not performed in a research lab and real-life data helps to better understand the needs of an athlete. Therefore, flexible, mobile und sufficiently accurate data recording is needed. Development Platform In a lab environment, highly accurate and comprehensive 3-D motion capture systems established as gold standard for movement analysis. However, for real-world data recording and processing, there is a high demand for mobile, lightweight, small and accurate solutions. We employ the capabilities of Body Sensor Networks (BSN) for data recording and live processing to provide research tools and develop innovative mobile applications. The key factors from an application perspective are usability, accuracy and price. The Pattern Recognition Lab uses BSN consisting of Shimmer Sensor nodes and Android-based mobile devices. The system is highly flexible as it can integrate different kinematic (Accelerometer, Gyroscope) and physiological (ECG, EMG) sensors and allows direct access to the unprocessed sensor data.