GSoC 2010/Projects/page-reader

(New page: So far, unpaper and tesseract are ported and I have managed to convert images taken by the camera to text and output them via scripts from the command line. My current task is to get the ...)
 
Line 2: Line 2:
...gotta go hit the road...will update in half hour...
...gotta go hit the road...will update in half hour...
 +
 +
== Optical Page Reader for the N900 - GSoC 2010 Project ==
 +
 +
*'''Student''': David Jordan ([[User:dmj726|dmj726]])
 +
*'''Mentor''': Florian Boor ([[User:florian|florian]])
 +
*'''Project Status''': accepted, in development
 +
*'''Repository''':
 +
** todo
 +
 +
Status updates for the project can be found on the blog: todo
 +
 +
===Project summary===
 +
 +
I plan to make an application for the Nokia n900 to allow blind and visually impaired people to read books, newspapers, magazines, signs, and other printed text on their own. The user simply opens the app and takes a picture of the document. The app then processes the image with an OCRFeeder-based backend and reads it aloud via text-to-speech software.
 +
 +
===Progress===
 +
 +
So far, unpaper and tesseract are ported and I have managed to convert images taken by the camera to text and output them via scripts from the command line.  My current task is to get the image processing working correctly, so that the spoken text is more accurate.  By the end of today, I should have this automated.
 +
 +
* porting except for ocrfeeder and two image processing steps - done
 +
* ocrfeeder needs pygoocanvas, which compiles but needs packaging
 +
* camera app is mostly there but needs a few tweaks.
 +
* GUI prototype is done, but needs some experimentation with controls.
 +
 +
Filed application for garage project, but haven't gotten approval yet.
 +
 +
===To-do===
 +
* Finish porting and packaging OCRFeeder.
 +
* Automate image processing.
 +
* Tie this to the GUI.
 +
 +
[[Category:Google Summer of Code]]

Latest revision as of 16:20, 16 July 2010

So far, unpaper and tesseract are ported and I have managed to convert images taken by the camera to text and output them via scripts from the command line. My current task is to get the image processing working correctly, so that the spoken text is more accurate.

...gotta go hit the road...will update in half hour...

Contents

[edit] Optical Page Reader for the N900 - GSoC 2010 Project

  • Student: David Jordan (dmj726)
  • Mentor: Florian Boor (florian)
  • Project Status: accepted, in development
  • Repository:
    • todo

Status updates for the project can be found on the blog: todo

[edit] Project summary

I plan to make an application for the Nokia n900 to allow blind and visually impaired people to read books, newspapers, magazines, signs, and other printed text on their own. The user simply opens the app and takes a picture of the document. The app then processes the image with an OCRFeeder-based backend and reads it aloud via text-to-speech software.

[edit] Progress

So far, unpaper and tesseract are ported and I have managed to convert images taken by the camera to text and output them via scripts from the command line. My current task is to get the image processing working correctly, so that the spoken text is more accurate. By the end of today, I should have this automated.

  • porting except for ocrfeeder and two image processing steps - done
  • ocrfeeder needs pygoocanvas, which compiles but needs packaging
  • camera app is mostly there but needs a few tweaks.
  • GUI prototype is done, but needs some experimentation with controls.

Filed application for garage project, but haven't gotten approval yet.

[edit] To-do

  • Finish porting and packaging OCRFeeder.
  • Automate image processing.
  • Tie this to the GUI.
Retrieved from "http://wiki.maemo.org/GSoC_2010/Projects/page-reader"