The explosion of the quantity and diversity of electronic documents in the last few decades has greatly contributed to the popularity of document image analysis as a computer vision research topic. Document image analysis involves the automatic extraction and interpretation of the information contained in documents destined for human comprehension, directly from the image format. In this presentation, we will present several research projects which are being conducted in the Computer Vision Research Lab at the University of Victoria in electronic document image analysis using computer vision techniques. More specifically, we will discuss document image segmentation and classification based on texture, the generation of a layered ground truth and a public business document image data base for the evaluation of algorithms, the development of an intelligent electronic library of business image documents and the interpretation of images of architectural plans.

Biography

Alexandra Branzan Albu completed her Ph.D. in Electrical Engineering at the Polytechnic University of Bucharest in 2000. From 2001 to 2003, she was a postdoctoral researcher in the Computer Vision and Systems Laboratory at Laval University. In 2004, she became an Assistant Professor in the GEL-GIF Department at Laval University. In 2005 she joined the Department of Electrical and Computer Engineering at the University of Victoria as an Assistant Professor and was then promoted to Associate Professor. Her research interests include computer vision, more specifically the practical applications of this discipline. Her research group is currently working on several projects, including electronic document image analysis, the analysis of mountain range landscapes in order to detect climate changes, salient event detection in underwater videos, etc. Her research is funded by industrial partners such as SAP Canada and by NSERC and MITACS.