[HASS-RG] Job vacancy at KCL
Tobias Blanke
tobias.blanke at kcl.ac.uk
Mon Oct 5 02:48:11 CDT 2009
http://www.kcl.ac.uk/depsta/pertra/vacancy/external/pers_detail.php?jobindex=8277
The Project
The Centre for e-Research at King's College London is seeking to appoint
a Software Developer to work on the OCRopodium project, a JISC-funded
project investigating the use of the open source OCRopus software
(http://sites.google.com/site/ocropus/) for applying Optical Character
Recognition (OCR) to historical and archival material.
The Centre has been involved in several collaborative projects that
digitised historical and archive material, and OCR is a key part of such
digitisation processes. However, in the past we have used proprietary
OCR software, which raised a number of issues: (i) commercial OCR
software and consultancy may be costly; (ii) the closed, “black box”
nature of commercial software is less flexible and adaptable; (iii)
digitisation staff at the university are de-skilled, as the OCR
expertise is concentrated in commercial hands; (iv) commercial OCR
software, which is typically developed for commercial applications, can
be inappropriate for historical and archival material.
The OCRopodium project aims to address these issues by:
• Trialling and evaluating an open source approach to OCR, using the
OCRopus software.
• Developing and training OCRopus components for historical and archive
material.
• Developing a centre of expertise in the use of OCRopus for historical
and archival material.
• Integrating OCR activities within semi-automated digitisation workflows.
The Role
The successful applicant will be the key technical staff member for the
project, and will be responsible for:
• Carrying out technical investigations into the functionality and
architecture of OCRopus. As OCRopus is an actively growing open source
project and thus imperfectly documented, this will in itself require an
ability to understand the source code and debug the software.
• Developing and integrating software components for OCRing historical
material, and enhancing existing components.
• Contributing new and enhanced components to the OCRopus open source
project.
• Benchmarking and evaluation of OCRopus, in collaboration with our
project partners at Queen's University, Belfast (QUB).
• Integrating OCR within broader digitisation and digital library workflows.
For an informal discussion of the post please contact Katrin Tiedau on
Katrin.tiedau at kcl.ac.uk
--
Dr Tobias Blanke
Research Fellow
Arts and Humanities e-Science Support Centre
Centre for e-Research, King's College London
26-29 Drury Lane, London WC2B 5RL
+44 (0)20 7848 1975
tobias.blanke at kcl.ac.uk
http://www.ahessc.ac.uk
More information about the hass-rg
mailing list