| KEYWORD |
Automatic Characters and Formulae Recognition @UniTO
keywords ASSISTIVE TECHNOLOGIES, BLIND PEOPLE INCLUSION, IMAGE PROCESSING, OCR
Reference persons PAOLO ERNESTO PRINETTO
External reference persons Prof. Anna Capietto (Università di Torino)
Dr. Nadir Murru (Università di Torino)
Giuseppe AIRO’ FARULLA (PhD candidate, Politecnico di Torino)
Research Groups TESTGROUP - TESTGROUP
Thesis type EXPERIMENTAL THESIS
Description Motivations:
Automatic recognition of both printed and handwritten characters remains a challenging problem in pattern recognition. Most existing Optical Character Recognition software (OCR) deal with it by exploiting simultaneously two highly correlated techniques: character segmentation and pattern recognition. As part of the OCR process, character segmentation techniques are applied to patterns representing individual characters to be recognized. State-of-the-art strategies for both characters’ segmentation and recognition unfortunately typically fail when considering mathematical formulae, graphs, handwritten and printed words with touching characters, representing well known issues that often occur in degraded or compressed text images (e.g., low-quality photocopies).
Conversely, the development of a reliable OCR software specifically tailored for scientific texts has been long required from the community of blind and visually impaired people, since only very few scientific texts are, to date, accessible.
Given that, such a software would represent both an improvement to the state-of-the-art and a valid assistive tool to blind and visually impaired persons addressing scientific studies.
Goal of the thesis:
The thesis aims at developing an automatic Optical Character Recognition (OCR) system specifically tailored for scientific text in order to assist blind and visually impaired persons during their studies.
Learning outcomes:
The candidate will acquire significant knowledge in the fields of image processing, code optimization, Assistive Technologies.
External/Industrial cooperations:
The research activities and the thesis are carried out in cooperation with:
- Università degli Studi di Torino, Department of Mathematics
- CINi National Lab on Assistive Tecnologies (https://www.consorzio-cini.it/index.php/en/national-laboratories/laboratorio-astech)
Required skills A background in image processing is a plus but not required.
Notes Number of required Students: 1 or 2
The research activities and the thesis are carried out in cooperation with:
- Università degli Studi di Torino, Department of Mathematics
- CINi National Lab on Assistive Tecnologies (https://www.consorzio-cini.it/index.php/en/national-laboratories/laboratorio-astech)
Deadline 05/10/2017
PROPONI LA TUA CANDIDATURA