Machine Vision Algorithms on Cadaster Plans

Platform Digital Humanities – Lecture Series

Monday, 4 December December 2017
16h30 – 17h30
Room R.218
Antwerp University City Campus

Cadaster plans are cornerstones for reconstructing dense representations of the history of the city. However, as some of these handwritten documents are more than 200 years old, the establishment of processing pipeline for interpreting them remains extremely challenging. The talk will present the implementation of an automated process capable of segmenting and interpreting Napoleonic Cadaster Maps of the Veneto Region dating from the beginning of the 19th century. The system extracts the geometry of each of the drawn parcels, classifies, reads and interprets the handwritten labels. This efficient and automated process opens new perspectives to reconstitute the past.

Attendance is free and open to all, but please register in advance.

About the speaker

Sofia Ares Oliveira received her MSc in Electrical Engineering from Ecole Polytechnique Fédérale de Lausanne (EPFL), Switzerland, where she specialized in signal processing. Since 2016, she has been working at the Digital Humanities Laboratory at EPFL. Her research focuses on developing computer vision and machine learning tools to process archival documents, which range from cadastral maps to handwritten manuscripts and printed documents.

About the lecture series

The series is an initiative of the platform{DH} at the University of Antwerp. The objective of the series is to sketch the evolving landscape of digital humanities. We found that the diversity of the field is perhaps best illustrated through concrete examples from practitioners. We invite speakers from a variety of disciplines, all united by the fact that they undertake exceptional and interesting research projects.