Aby gets digital: ARAby -An adaptive Retrieval- and Analysistool for the support of image-oriented scientific research

• Applicants: Professor Dr. Ralph Ewerth; Professor Dr. Eyke Hüllermeier; Professorin Dr. Eva-Maria Seng
• Subject Area: Image and Language Processing, Computer Graphics and Visualisation, Human Computer Interaction, Ubiquitous and Wearable Computing ; Art History
• Term: from 2017 to 2018
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 389247364

The project Aby gets digital: ARAby. An adaptive Retrieval- and Analysistool for the support of image-oriented scientific research describes the proposal of the development of an Software-Tool for the digitized research process in the humanities. It aims to improve the efficiency of Image-Retrieval in databases to enhance the analysis of image data within scientific work in order to facilitate new theory formation. The tool tends to increase the performance of national and international scientific information systems. The original idea is based on Aby Warburgs method of comparing and analyzing pictures out of different contexts simultaneously together. For his so called Mnemoyne Bilderatlas he arranged pictures on wooden panels covered with canvas in an ephemeral way and constantly re-arranged an edited his selection. The resulting compositions of images were as well a working method as a publication of results and they allowed innovative new research questions. The e-Research-Tool ARAby intends to carry Warburgs procedure forward to the digital age and even to expand it with current information technology. The analysis of digital images based on the criterion of similarity should be undertaken by ARAby in an automated way, complemented by additional functions. Essential for the implementation of ARAby are the three functionalities pattern recognition, clustering and (user)adaption. One fundamental function is the pattern recognition of pre-defined concepts and the annotation of the resulting information within the metadata of the images. Search queries can be made by keywords (text) as well as content based (Image). The results of queries processed by ARAby are then displayed in a particular form of ranking, in which the user analysis feature is integrated. As an e-Research-Tool, ARAby should enable scientist in the humanities to search huge (image) databases by using their common browser on the web. Standardized interface functions are taken into consideration to obtain transferability to other databases. ARAby should enable easier image retrieval for scientist by using comprehensible, meaningful and transparent ways of displaying the results.

• DFG Programme: Research data and software (Scientific Library Services and Information Systems)
• Cooperation Partner: Professor Dr. Gilbert Lupfer