Re: [orca-list] ocr

[Willem van der Walt <wvdwalt csir co za>]

Hi,
I had the ABBYY finereader for linux working.
One of my upgrades seem to have broken it, but I did not need much OCR 
lately.
It was 12 thousand pages per year, and I used it for quite a few years.
The one I had was a command line tool which is fully accessible.
My employer bought it at the end of 2011, almost ten years ago.
Here is the link we used at the time, but it did not work just now  when I 
tried it.
https://store.abbyyeu.com/cgi-bin/shop/curr=EUR/ml=EN?ID=FRE9LIN12KPPY
Cuneiform works ok.
One can scan a book, both pages at once. Cuneiform then decolumnizes it 
correctly.
Tesseract is a lot better than it used to be, but it is sencitive to the 
image you feed it.
HTH, Willem


On Mon, 1 Mar 2021, Francisco Tissera via orca-list wrote:

> Hi there.
>
> Cuneiform is also a thing, and I believe it does layout annalisis as well, idk about it’s accessibility, but 
> I think
> ocrfeeder and lios both support it. I also believe  it’s accessible from a cli interface.
>
> From: Rastislav Kish via orca-list
> Sent: Monday, March 1, 2021 8:55 PM
> To: orca-list gnome org
> Subject: Re: [orca-list] ocr
>
>  
>
> Hey there,
>
> so, as others already mentioned, there is Tesseract, now also with its neural networks based models.
>
> Although I have to admit, that I'm kind of lost in its versions and models. I have installed data from the 
> Ubuntu's
> repository, but somewhere I've read those are not the bestones, but rather a compromise between quality and 
> speed.
>
> So, IDK.
>
>  
>
> There is also FineReader engine, which seems quite complex.
>
> Although I wonder, does anyone here actually have it?
>
> I've read somewhere something about 10000 pages limit after buying a licence, but I have no idea whether it's 
> true for home
> users or not.
>
> Abbyy support is one of the worst I've ever seen, may be I just didn't have luck, but they did not answer 
> single of my
> emails.
>
>  
>
> The price is also quite hidden. I recall something about $180, but I was not able to find it when I last 
> checked. Their site
> is quite a drag.
>
>  
>
> But the quality is good, that's true.
>
>  
>
> Best regards
>
>  
>
> Rastislav
>
>  
>
> Dňa 1. 3. 2021 o 8:53 mattias jonsson via orca-list napísal(a):
>
> maybe this has been asked before
> on win we have omnipage,abbyy fine reader etc
> whathave linux?
>
>