[Python-apps-team] Thank you for offering FOSS OCR to us

[Susmita/Rajib Bandopadhyay]

Sirs,
Thank you once again to kindling the hope for freedom from the Proprietary
OCR software, particularly, ABBYY finereader.
Sirs, I have tested OCRfeeder and find that some functionality and
interactivity needs to be added, like:

(1) The OCRed text be output onto an editable Wordprocessor rather than
plain text editor with option to
(a) map each block area to a separate textbox at same absolute position.
(b) Allow each block to have the option of spell-checking manually.

(2) Different blocks for image, plain text, formatted text and tables.

(3) Use of only keyboard to incorporate spellings and editing without
lifting fingers from the keyboard with another interactive window with
focus upon it.

(4) Abbyy has a *foolishness* that we *wouldn't* want *us* to *repeat*. *If
you feed a pdf file to Abbyy it converts the characters into images and
then OCRs the image and re-converts them into text, thus incorporating
errors.*

Then we would be proudly say that our open source OCR is the best in the
world!

Alas, sirs, I am a seft-taught computer buff, and am learning scripting and
programming in c. I am very sad not to be able to help you with codes! :-(

Sirs, please consider my suggestions as humble prayer.

Regards,
Rajib Bandopadhyay
Your admirer


On 04/06/2015, Rajib Bandopadhyay <bkpsusmitaa at gmail.com> wrote:
> Dear Sir,
> Please don't thank me for using OCRFeeder. Rather, it is your values
> that gave us the product for free.
> Sir, among all Linux systems I tested, knoppix is the least
> CPU-hungry, for an Indian laptop which is Defective By Design. It has
> a NVidia GPU, and unless used, rather than the CPU, the system heats
> up and shuts itself down. Even Ubuntu, despite with NVidia drivers
> installed heats up the system.
> Unfortunately, only Knoppix 7.2.0 is compatible with the NVidia
> drivers. The latest Knoppix isn't. Also, the Debian Repo of 7.2.0 has
> been deprecated raising the difficulty level for me installing the
> package via synaptic.
> Sorry to have bothered you with unnecessary detail.
> Thank you, Sir, on behalf of FOSS community. One day I am going to use
> OCRFeeder with the latest Tesseract and other engines exclusively.
> Regards,
> Rajib Bandopadhyay
>
> On 04/06/2015, Joaquim Rocha <me at joaquimrocha.com> wrote:
>> Hi Rajib,
>>
>> Thank you for using OCRFeeder.
>>
>> Tesseract is automatically detected if it's already installed in your
>> system. As for installing it, I think that it's hard to be installing
>> all the dependencies by hand. Why are you using Knoppix? Can't you use
>> Ubuntu with a live USB/CD, it should be essentially the same but Ubuntu
>> should have the most recent packages.
>>
>>
>> Best regards,
>>
>> --
>> Joaquim Rocha
>> http://www.joaquimrocha.com
>>
>> On 02/06/15 20:57, Rajib Bandopadhyay wrote:
>>> Dear Sir,
>>> I used Abbyy Finereader from time to time obtained when I bought a
>>> scanner and therefore I was dependent on Windows. I used a virtualised
>>> windows for it.
>>> I tried to wean myself away from the package, but there was no
>>> comparative OCR like Abbyy's.
>>> Thank you for the software.
>>> Having said this, how do I include the tesseract engine to your
>>> software? I have a snapshot but I don't know what to do, and from
>>> https://wiki.gnome.org/action/show/Apps/OCRFeeder I find that the help
>>> page is in German, but Google translation helps the first para.
>>> I have a Knoppix (Debian) system and I installed the package from the
>>> above link. In my knoppix I can't use aptitude. It is an old version..
>>> So I downloaded the tesseract .deb file to use gdebi, but it has
>>> several dependencies. How do I use tesseract, or any other engine from
>>> scratch without synaptic?
>>> Eager to hear from you,
>>> Rajib
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.alioth.debian.org/pipermail/python-apps-team/attachments/20150715/cf031fb0/attachment.html>