Ubuntu QA:
BlogBrainstormPackage status
Log in
Ubuntu QA
Evince Document Viewer
Idea sandbox Idea sandbox
Popular ideas Popular ideas
Ideas in development Ideas in development
Implemented ideas Implemented ideas

Contributor Lucky LIX on Evince Document Viewer

Universal document reader  
Written by cousteau the 15 Feb 09 at 18:28. Not an idea
Sometimes you open a MS Word .doc or a OOo Writer .odt document in order to read it and have to wait for OpenOffice.org to get loaded. Once it has loaded, you get a blinking cursor and a lot of edition buttons and tools (cut, paste, bold, italics, font, font size, format stuff...)

This can be annoying when you just want to read a document. It's like opening all images with The Gimp instead of an image viewer (EoG, gThumb).
836
votes
closed
Solution #1: Make Evince read more formats (ODF, Office, etc)
Written by cousteau the 15 Feb 09 at 18:28.
Evince should be able to read more formats, like ODF (.odt), Office (.doc), and optionally plain text (.txt, .log...) and HTML (.htm, .html).

Since Evince can provide thumbnails for Nautilus, this would also extend the number of thumbnailed files.
290
votes
closed
Solution #2: Add an edit button
Written by deathsshadow77 the 16 Feb 09 at 02:22.
Do the same as above but add an edit button to evince
45
votes
closed
Solution #3: Also add conversion&export to mobile devices option
Written by Dinth the 16 Feb 09 at 16:49.
Like exporting to Mobipocket format in upcoming KDE4.3 version of Okular.
316
votes
closed
Solution #4: Add "Quick Look" style preview in file browser
Written by belovedmonster the 16 Feb 09 at 18:22.
Apple has a great feature in its file browser where you hit the spacebar and you can view documents/pdfs etc in fullscreen, but clicking on the file as usual loads it up in its usual application.

See it in action here
http://www.youtube.com/watch?v=ti9NehCxhDQ

It supports plugins so basically anyone can add more supported file formats.

I would love to see something similar in Ubuntu.
39
votes
closed
Solution #5: Help Gloobus Developer
Written by BadChoice the 19 Feb 09 at 07:14.
Gloobus is a preview application that now supports PDF it would be great if it could also show openoffice documents, words and comic books, just for viewing, they all can easely be developed as plugins but, well, the plugins need to be developed :D
99
votes
closed
Solution #6: File manager plugin support
Written by Ivo Georgiev the 28 Feb 09 at 14:22.
Make the file managers support universal plugins for previewing/reading media files. For example, when you install Evince, a plugin is installed for the file manager to support previewing PDF files. The plugin is used by Nautilus, Dolphin and Thunar or other file managers (such as PCManFM).

The plugins should also have information inside them about
where there work best. For example, the Evince plugin has information in it that it works best in Nautilus, so every file manager chooses the plugin that works better in it for every filetype.

If there is only one plugin for this filetype, it should be used no matter in which file manager it works best.

Also, the plugins are used if there are usable, so the plugin for audio files can be located in the Nautilus package, but use mplayer. And the plugin should be only enabled if mplayer is available.

This way it should be possible to create a plugin for reading office files using the OpenOffice.org framework
without starting the whole program.

See the 6 comments or propose a solution (latest comment the 5 Aug 11 at 19:25) >>

There is no easy GUI solution for OCR in Ubuntu.  
Written by hunt.topher the 30 Mar 09 at 02:15. New
OCR (Optical Character Recognition) is a useful tool for anyone who has PDFs or other documents containing scanned verbal material that you want in copiable text form (for storage purposes, for use on a portable reader device, etc).

Anyone who wishes to take advantage of the OCR tools available in Linux to convert scanned PDFs into text must generally rely on command line tools (for converting the PDF into images, for converting images into the right format, or for running the OCR program) to get the job done. This is an effective barrier to use for the average office worker.
172
votes
up equal down
Solution #1: Incorporate OCR capabilities into Evince
Written by hunt.topher the 30 Mar 09 at 02:15.
Given that open-source tools are available to fulfill this function (Imagemagick to convert PDFs, Tesseract to OCR to plain text), it would be useful to have a GUI button in Evince to output text from a scanned PDF.

A button "Convert this document to text" could convert a PDF into the correct image format and run an OCR program such as Tesseract to produce text, then display that text in Text Editor, all from one button-click.

Perhaps this could begin as an optional plugin while under testing.
113
votes
up equal down
Solution #2: Incorporate OCR capabilities into OpenOffice
Written by Darwin Survivor the 30 Mar 09 at 16:52.
I think it would be more useful to have this in OpenOffice (file import > pdf via ocr). Not only could we edit it from there, but OpenOffice can export directly to pdf.

I don't think we should add this to evince, because evince is a nice "light" pdf reader, and should stay that way. OpenOffice on the other hand is an office suite which already exports to pdf.
70
votes
up equal down
Solution #3: Use gscan2pdf
Written by oliver-joos the 31 Mar 09 at 15:27.
Try the latest gscan2pdf (>= 0.9.27). It has a Gnome GUI and is nice to scan and reorder multi-paged documents. For OCR it uses Tesseract or GOCR (try 300dpi and Tesseract).

To further improve recognition on grey/old paper or with coloured text I tweaked it a bit: gscan2pdf uses "unpaper" to clean text-pages before OCR, which IMHO does not lower error rate significantly. I replaced "unpaper" with a script that calls "convert" from "imagemagick" mainly to "-contrast-stretch", with impressive results!

What you cannot do with gscan2pdf is OCR of pages with complex layout. (multiple columns, tables, ect.)
21
votes
up equal down
Solution #4: use Ocropus+Tesseract
Written by JuliusH the 9 Apr 09 at 01:46.
develop a gui for Ocropus and make it the default ocr-app for ubuntu
0
votes
up equal down
Solution #5: Use the Java-GUI jtOCR
Written by vhindriksen the 9 Sep 09 at 16:44.
See the comments how to get it. It should get fixed for Ubuntu and get into the repositories, just like solution 3 and 4. No defaults, just choices.
3
votes
up equal down
Solution #6: EASY-OCR
Written by nalin4linux77 the 7 May 11 at 03:00.
EASY-OCR-2.5 (WITH 24 LANGUAGE SUPPORT)

Now a visually impaired person can read print in 24 languages using free software. new features.

1. being a deb package easy-ocr 2.5 can be installed very easily.

2. scan and read from very beginning.

3. settings are saved until you go for change of settings again.

4. file name and location is requested by the programme at the beginning of the scanning process, if no location is specified file will be automatically saved in the documents.

5. auto rotation. now, you are no more to worry about how you keep the book on the scanner. programme can set the correct rotation for you.

6 Two engines. there are two engines one good for picture skipping and speed and the other for lay out analysis.

7 repeated scanning. now there is fecility for repeated scanning and one can stippulate what should be the delay between the scanning.

8. page number. programme will automatically give the page number and one can go to the page using find fecility.

9. uninstallation. one can uninstall the programme by apt-get remove easy-ocr.

For 11.04 users please download scribes from here http://packages.ubuntu.com/maverick/all/scribes/download
Please send suggestion and problems to nalin4linux77@gmail.com
HOW TO USE EASY-OCR

1. after installation, please go to the graphic menu and select settings or press alt+ctrl+shift+s to adjest the scanner settings.now the programme will announce if it has detected the scanner, now one should adjest the delay between repeated scanning, resolution, angle in which you have kept the book, brightness, language. after selection programme will announce the settings you have selected.

2. regarding auto rotation,one can select either manual settings or automatic method. in the automatic mode one has to type one possible word in the selected page (example, the, is, to) and then press return. programme will automatically select the correct rotation.

3.now one can start scanning. there are two engines, select easy-ocr-scan1 or 2 in graphics menu or press alt+ctrl+shift+1 to start scanning with the first engine which is good for picture skipping and speed. press alt+ctrl+shift+2 for working with the second engine which is good for layout analysis.

4. reading key. after the first scanner movement and recognition , espeak will announce 1 and now press add button in the numpad and add button in the numpad to start reading.

5. alt+ctrl+shift+c will stop or cancel scan at any time.

6. you can go to any page by pressing ctrl+f and typing the following, page-number of the page. blank pages will be skipped by the programme.

EASY-OCR is made as user friendly as possible. you can make it more friendly through your suggestions. please contact the following emails. sath.linux@gmail.com and nalin4linux77@gmail.com
Please send suggestion and problems to nalin4linux77@gmail.com
1
votes
up equal down
Solution #7: linux-intelligent-ocr-solution
Written by nalin4linux77 the 27 Feb 12 at 05:59.
LIOS-1.2

LIOS is a free and open source software for converting print in to text using either scanner or a camera. It can also produce text out of scanned images from other sources. Program is given total accessibility for visually impaired. LIOS is written in python and we release it under GPL3 license. LIOS will work with Debian based operating systems. LIOS is an effort from the easy-ocr development team. There are great many possibilities for this program. Feedback is the key to it. expecting your feedback. nalin4linux77@gmail.com and sath.linux@gmail.com.
HOW TO INSTALL

Download deb file from here http://linux-intelligent-ocr-solution.googlecode.com/ open it and install
What is new in LIOS-1.2
1 Cam-Scan,
2 Cam-Reader,
3 Scan-to-image-only,
4 Scan-to-images-repeatedly,
5 Introduction of py-sane, Glaid library make the program faster and efficient,
6 Multiple arguments are handled effectively,
7 Ocr a single Image,
8 Artha shortcut (alt+control+W),
9 Beta version of spell-checker,
10 Provision for submitting issues in the About Dialog.
Features
1 Single scan & Repeated Scanning,
2 Ocr Folder,
3 Ocr Pdf,
4 Ocr image only,
5 Cam-Scan and Cam-Reader,
6 Scan-for-image-only & repeatedly,
7 24 Language support (Given at the end),
8 Full GUI environment,
9 Selection of starting page number, page numbering mode and number of pages to scan,
10 Selection of Scan area, brightness, resolution and time between repeated scanning,
11 Full Auto Rotation,
12 Brightness optimizer,
13 Audio converter,
14 Easily Accessible Preferences Window,
15 5 OCR Engines (OCROPUS,CUNEIFORM,TESSERACT,GOCR,OCRAD),
16 Good text manipulation with Find, Go-To-Page, Go-To-Line, Append file, Punch File.
17 Display Preferences for Low vision,
18 Dictionary Support for English(Artha)
19 Beta version of spell-checker,
20 Provision for submitting issues,
21 And more features are in the preferences.
How to start using LIOS.
1. Scanning.

In order to start new scan, first press ctrl+n and then press f9 for single scan or ctrl+f9 for repeated scanning. To set the scanning preferences press ctrl+p and set the starting page number, Mode of page numbering, double page mode if you intend to keep 2 pages at a time, rotation to select the way in which you want the program to rotate the images before conversion. In full automatic rotation mode, one can keep the book in 00 90 180 and 270 degree angle. In partial rotation mode program will scan once to find out the position of the book and then the rotation will be kept. In manual mode one should select the angle. partial and manual mode is faster than full auto rotation mode in ocr process. One can select the number of pages to be scanned at a stretch by setting number of pages in the case of repeated scanning. One can stop all scanning process by pressing ctrl f4.
2. Cam-scan.

one can now use Hovercam or a Webcam to produce text in LIOS. Adjustments with these devices can be made using LIOS-cam-preferences in edit menu. This feature will help to read books and other printed materials such as visiting cards currency and like and also it makes the ocr process very fast and accurate. Please be specific to use devices with auto focusing facility. remember that there is no autorotation in this utility.so for the same reason, support of a stand for the webcam will be highly appreciated.
3. Cam-reader.

is the utility which will give a continuous output as one moves the webcam. First it will create the image and then will produce the text and it will start reading. After the completion of reading, it will repeat the process automatically. In cam-scan, one has to take the photo and it will be converted in to text.
4. Ocr Image.

LIOS can convert image file to text which is in jpg, tif, png, pnm and bmp.
5. Ocr folder.

LIOS can convert scanned images from other sources. It can convert jpg, jpeg, tif, tiff png, pnm, formats. To convert the images in a folder, select scan from folder option from scan menu and then select the input folder.
6. Ocr Pdf file.

Select Ocr pdf from scan menu and then select the input file. It is recommended that one can use ocropus as engine more efficiently in pdf conversion.
7. scan for image only and scan for images only repeatedly.

Help one to scan only images and it will give the user opportunity to utilize different ocr engines conveniently. Also it avoids delay between each scan if one does not want to listen to the output. Images will be saved in LIOS or one can choose his own destination. Now conversion can be done using folder option.
8. Brightness checker.

To set a n exact value of brightness or threshold is the best way to ensure maximum efficiency out of ocr engines. To find out the best value, go to tools menu and select brightness checker. This utility will scan for 15 or 17 times to complete the process. After the process, number of words detected at different values will be shone in tabs. If you want to

See the 8 comments or propose a solution (latest comment the 3 Feb 11 at 12:50) >>