[maemo-developers] Project GSoC 2009 -- Text optical recognition

[From: Marcio Miranda]

Hello everyone,
I am new here and I'm posting my proposal for GSoC 2009, suggestions
are welcome.
Can anybody tell me how to get in contact with potential mentors?
Thanks


* Project title:
PATROPY -- Text Optical Recognition


* Project Abstract (short description):
The proposed programme of acquisition of text and recognition of
characters, is to use the low-resolution cameras of equipment N800 to
capture images and through the manipulation of its contents identify
characters.


* Project full description:
This project has started in July 2008 but was some time without
manuntenção. Was initially characterized as only one port of a
solution already exists for optical character recognition (OCR) for
maemo platform. However this new approach seeks to
desenvonver/implement a new structure, comprising: an algorithm (state
of the art) of OCR recognition, a more friendly interface with more
options for setting and passage of the algorithm parameters, and
better integration and system .


* Project tasks and deliverables:
01 - lift the state of the art algorithms for OCR (relieve capacity of
the device);
02 - implement the algorithm in the simulator (SDK);
03 - to develop the user interface (HCI);
04 - to integrate the interface with the algorithm;
05 - set the application infrastructure to the device.


* Project schedule:
April 20th to May 05th:
Get to know my mentors, achieve the lifting of the state of the art
OCR algorithms.

May 05th to June 05th:
Implementation in C of the algorithm chosen in the OCR task before.

June 05th to July 05th:
PyGTK implementation in the interface to use the PATROPY, integrate
with the algorithm of OCR recognition;

July 05th to July 20th:
Packing and integrate with the infrastructure of the device (MAEMO
architecture).

July 20th to August 17th:
Testing and Documentation work.

August 18:
If successful - partying too:)

	
* Do you have other commitments during the Summer of Code coding
period? If yes, what kind of commitments?
No, I do not have.


* Why are you the right person for this project?
Why work on my Masters Thesis in image processing, and last year I
have started a project with the same subject available for the GSoC
2009. For more information:
https://garage.maemo.org/projects/patropy/.


* What are the benefits of your project for the Maemo Community?
Allow people with visual difficulties, or even people with different
nationalities can interact with each other through the recognition of
text and possibly (plugin) translation of these texts to another
language!


* Short Biography (academic work, open source background and other
activities that you like to do)
I am a student of Masters in the Federal University of Parana, my
research is related to detection of events related to security in the
videos. I have worked in recognition of characters, and I have
experience with the platform MAEMO. I stock for approximately 4 years
in C3SL (http://c3sl.ufpr.br) center of excellence in free software in
Brazil.
My work and articles are listed below:

01 - MIRANDA, M. R.; MONTANO, R.; SIIVA, L. F., Castilho, M.; HEXSEL,
R.. Business Intelligence in the Public Schools of the State of
Paraná. In: WFS 2008 (Workshop for Free Software), 2008, Porto Alegre
/ RS. WSL 2008 (Workshop for Free Software), 2008.

02 - MIRANDA, M. R.; MEIRA, J. A.; VIGNATTI, T. . Self-recovery on
Linux systems. In: Regional Conference of Undergraduate Science and
Technology in Engineering, 2004, Curitiba. CRICTE 2004, 2004.

03 - MIRANDA, M. R.; MEIRA, J. A.; PICUSSA, J.. Simplified system of
administration of users. In: Evince, 2006, Curitiba. 14 - Evince -
Event of Scientific Initiation. Curitiba: Imprensa Universitária da
UFPR, 2006. p. 255-255.

04 - MIRANDA, M. R.; MEIRA, J. A.; VIGNATTI, T.. Self-recovery on
Linux systems. In: Evince, 2005, Curitiba. 13 - Evince - scientific
initiation event. Curitiba: Editora da Universidade Federal do Paraná,
2005.

05 - MIRANDA, M. R. ; PICUSSA, J. , SCHMITT, E. V. . Course
administration and use of Linux for the Paraná Digital project. 2004.
(Course of short courses / other).

06 - MIRANDA, M. R. ; PICUSSA, J. , SCHMITT, E. V. . Apostila Paraná
Digital Project. 2004. (Development of instructional or teaching
material - material aid to the administration and use of Linux
operating system).

Best regards
Marcio


-- 
Marcio Roberto Miranda
Bacharel em Ciência da Computação.
Integrante do Programa de Pós-Graduação em Informática da UFPR.