Software to extract text embedded in images?
Thread poster: Pablo Bouvier
Pablo Bouvier
Pablo Bouvier  Identity Verified
Local time: 11:07
German to Spanish
+ ...
Sep 30, 2010

Please, can someone tell me if there is a program to extract text embedded in images, which allows to translate and to embed thetranslated into the same image? I remember reading that there is a program that allows this, but I'm noty able to remember the name of the programm. Thank you very much everybody!

[Edited at 2010-09-30 21:05 GMT]


 
mediamatrix (X)
mediamatrix (X)
Local time: 05:07
Spanish to English
+ ...
Curious question... Sep 30, 2010

There must be at least as many solutions to this problem as there are image file formats. Or, maybe, as few solutions as there are image file formats which embed text in a way that makes it reliably distinguishable from other image content when analysed by software.

MediaMatrix


 
Romeo Mlinar
Romeo Mlinar
Portugal
Local time: 10:07
English to Serbian
+ ...
OCR app for the first part of the question Sep 30, 2010

...such as ABBYY Fine Reader. And for superimposing the translation back on the image - I'm not sure it's that simple. You need precise editing, or, if the images are simple an uniform, a script can be written to put characters into the files/images. But, this is a complex task for an average translator without programming skills.

Somebody else might have the solution.


 
D@ve
D@ve
Germany
Local time: 11:07
English to German
+ ...
Not possible... Sep 30, 2010

Yep, Fine Reader is (imo) one of the best OCR Programms. To transfer the translation back to the image is theoretically impossible.

1. You need the font to write the text again
2. The Software needs to know what is "behind" the letters to recreate the background

Guess you have a word "MMM" in the Image and the translation would be "lll". The translation don't need as much space as the original. If you have black letters on white background that wouldn't be a pro
... See more
Yep, Fine Reader is (imo) one of the best OCR Programms. To transfer the translation back to the image is theoretically impossible.

1. You need the font to write the text again
2. The Software needs to know what is "behind" the letters to recreate the background

Guess you have a word "MMM" in the Image and the translation would be "lll". The translation don't need as much space as the original. If you have black letters on white background that wouldn't be a problem, but if you have an Image as background the software would have to recreate the background.
I use Photoshop for these kind of jobs (still need the font).

regards, Dave
Collapse


 
Soonthon LUPKITARO(Ph.D.)
Soonthon LUPKITARO(Ph.D.)  Identity Verified
Thailand
Local time: 16:07
English to Thai
+ ...
PDF editor Oct 1, 2010

For the second part, a PDF editor can do well [some PDF editors are freeware!]. The working steps are 1. Use OCR to convert image into texts 2. Use CAT (or none) to translate the texts 3. Use the editor to write back translation cleverly on layouts of the original image. Of course, your original image must be converted into a PDF file (e.g. by using Adobe Acrobat Professional) to work under these steps.

Soonthon Lupkitaro


 
John Fossey
John Fossey  Identity Verified
Canada
Local time: 05:07
Member (2008)
French to English
+ ...
Paint.NET Oct 1, 2010

For very limited use - a small handful of texts in an image - I sometimes use the free program Paint.NET to cut out the original text in an image and create new text. But it's time consuming manual work and I only really do it as a service as part of a regular document.

 
Jaroslaw Michalak
Jaroslaw Michalak  Identity Verified
Poland
Local time: 11:07
Member (2004)
English to Polish
SITE LOCALIZER
Quite possible Oct 1, 2010

Dave Remmel wrote:
To transfer the translation back to the image is theoretically impossible.


Apparently the good folks at ABBYY did not know this, as FineReader does it semiautomatically... That is, it extracts the text from the picture and fills in the gaps with the surrounding background color.

Of course, this works well mostly with uniform backgrounds (e.g. text pasted on a colorful photo is out of question), but it works.

If you want to try something more sophisticated, you can have a look at InPaint:

http://www.theinpaint.com/

It does a very good job of removing foreground objects from pictures... Adding a new text after that is rather trivial.


 
Pablo Bouvier
Pablo Bouvier  Identity Verified
Local time: 11:07
German to Spanish
+ ...
TOPIC STARTER
Software to extract text embedded in images? Oct 1, 2010

Jabberwock wrote:

Dave Remmel wrote:
To transfer the translation back to the image is theoretically impossible.


Apparently the good folks at ABBYY did not know this, as FineReader does it semiautomatically... That is, it extracts the text from the picture and fills in the gaps with the surrounding background color.

Of course, this works well mostly with uniform backgrounds (e.g. text pasted on a colorful photo is out of question), but it works.

If you want to try something more sophisticated, you can have a look at InPaint:

http://www.theinpaint.com/

It does a very good job of removing foreground objects from pictures... Adding a new text after that is rather trivial.


Hi Jabberwock: Thanks a lot for sharing the reference of InPaint.
This seems to be very near of what I am looking for or it is.

[Edited at 2010-10-01 09:12 GMT]


 


To report site rules violations or get help, contact a site moderator:


You can also contact site staff by submitting a support request »

Software to extract text embedded in images?







CafeTran Espresso
You've never met a CAT tool this clever!

Translate faster & easier, using a sophisticated CAT tool built by a translator / developer. Accept jobs from clients who use Trados, MemoQ, Wordfast & major CAT tools. Download and start using CafeTran Espresso -- for free

Buy now! »
TM-Town
Manage your TMs and Terms ... and boost your translation business

Are you ready for something fresh in the industry? TM-Town is a unique new site for you -- the freelance translator -- to store, manage and share translation memories (TMs) and glossaries...and potentially meet new clients on the basis of your prior work.

More info »