- Definition and Evaluation of the NEOCR Dataset for Natural-Image Text Recognition (2011)
- Recently growing attention has been paid to recognizing text in natural images. Natural image text OCR is far more complex than OCR in scanned documents. Text in real world environments appears in arbitrary colors, font sizes and typefaces, often affected by perspective distortion, lighting effects, textures or occlusion. Currently there is no dataset publicly available that covers all aspects of natural image OCR. A comprehensive well-annotated configurable dataset for optical character recognition in natural images is defined and created for the evaluation and comparison of approaches tackling with natural-image text OCR. Furthermore, current open source and commercial OCR tools have been analyzed in various test scenarios using the proposed NEOCR dataset. Based on the results further steps to be addressed by the OCR community are concluded towards all-embracing natural-image text recognition.
- Klimawandel. Mediengesellschaft (2012)
- Wissenschaftler der verschiedensten Wissenschaftsdisziplinen haben bei den Erlanger Universitätstagen 2010 und 2011, einer jährlich in Ansbach und Amberg abgehalten Vortragsreihe, zu Themen aus den Bereichen Klimawandel und Mediengesellschaft mit deren Vielzahl von Facetten Stellung genommen.
- The Erlangen Glaucoma Registry: a Scientific Database for Longitudinal Analysis of Glaucoma (2011)
- Background: Due to the slow progression of the Glaucoma disease, a large study population and long-time observations are needed to gain insights into its long-term effects and progression rates. Since modalities can export data in machine-readable formats, statistical analyses of the large number of examinations is feasible. These data have been integrated in a central patient registry, the Erlangen Glaucoma Registry (EGR). Objectives: The primary focus of the EGR system design has been its fitness for analyses. It holds almost all the available research data for registered glaucoma patients. This allows for cross-sectional and longitudinal observations and for evaluation of prognostic validity of diagnostic procedures. Methods: An adequate technology for integration of data and flexibility in data analysis is a database management system (DBMS). Here, a careful schema design is mandatory. Adding new modalities leads to schema modifications which are supported by defining a core database schema and attaching all data to this core. On that basis, a large number of modalities have been connected to the EGR. Results: The registry contains data of 1,400 patients in the main longitudinal study. It has successfully helped in scientific research, as can be seen in a large number of published papers. For example, validation of new sensory physiological methods requires patients with reliable diagnoses. The existence of a well-documented patient collective facilitated finding such patients. Conclusions: The EGR holds a unique amount of available data gathered in large longitudinal studies. It was successful in terms of medical results obtained. Developed as an evolutionary system, it can easily be extended.
- A Flexible, Adaptive System for Data-Stream Processing in Energy-Constrained Ad-hoc Networks (2015)
- Today's generation of sensor networks reflect several changes in system characteristics over traditional sensor networks. These changes affect three key aspects: energy as a fundamental resource, stream data processing, and the inherently dynamic structure of the overall system. In this paper we extract and present eight distinct challenges aligned to the key aspects which need to be addressed in the future. We use data of ongoing science and research projects to extract the most important challenges. These challenges need to be tackled in order to provide the basis for a flexible and adaptive system design which supports data-stream processing in energy-constrained ad-hoc networks.