Developing customized web applications integrating visual and flexible workflow using the Oppidum framework.

During this 7 months, I integrated a research project dealing the Vacheron Constantin archive. In this context, I developed successfully a classifier discriminating handwritten segments from printed ones, a new transcription system based on Hidden Markov Models, another one using recurrent neural networks (Long Short-Term Memory model). I also contributed to the development of the web viewer of the project. During this period,I also contributed to the Venice Time Machine project, an international flagship project involving several European universities, notably for defining data publication strategies and organizing scientific challenges.

In this project, I focused on the automatic transcription and textual alignment of 13th century Venetian document. In this context, I developed a recognition and alignment software using Hidden Markov Model and I co-developped a web interface for integrating this recognition system.

Project title: Writer Identification and Text Recognition Systems for Historic German Handwriting

Project titles:- A Study on Printed Multi-Font and Multi-Size Arabic Word Recognition at Ultra-Low Resolution- A New Arabic Printed Text Image Database and Evaluation Protocols (https://diuf.unifr.ch/diva/APTI/)