All the issues i see around this subject are old but i still had many
issues setting up tesseract on an aws ec2 instance.
$TESSDATA_PREFIX is set to /usr/local/share and the eng.traineddata is
present in the tessdata ( /usr/local/share/tessdata) folder along with the
config files.
Can someone pl
Hi,
my name is Giuseppe, i have this problem. I use tesseract-ocr 3.05 with
shell integration, but is really slow.
i haven't find the way to use it by dll, is this possible? The project
unfortunatly must be developed on vb6 platform.
Thanks
--
You received this message because you are subscribe
See answer inline.
Il giorno ven 3 mag 2019 alle ore 03:48 Tairen Chen
ha scritto:
>
> 1. I define the "--max_iterations 2" but the training stops at
> 5700, like below:
> " At iteration 351/5700/5700, Mean rms=0.117%, delta=0%, char
> train=0%, word train=0%, skip ratio=0%, wr
>There are three model sizes: best, normal and fast. Each of these can also
be converted to an integer model.
Only `best` can be converted to integer and in fact the LSTM models in
`tessdata` are the integer versions of best along with the base/legacy
models.
`fast` models have been trained with
Shree, thanks for the clarification.
Il giorno ven 3 mag 2019 alle ore 11:59 Shree Devi Kumar <
shreesh...@gmail.com> ha scritto:
> >There are three model sizes: best, normal and fast. Each of these can
> also be converted to an integer model.
>
> Only `best` can be converted to integer and in fa
I need to compile, by the way I use *OpenCL*.
My roadact on Ubuntu:
sudo apt-get install opencl-headers
sudo apt install ocl-icd-opencl-dev
export LDFLAGS=-L/usr/lib/x86_64-linux-gnu
./configure --enable-debug --enable-opencl
...
Configuration is done.
You can now build and install tesseract by ru
First of all: 3.04 is old. If you want to use 3.x tesseract use 3.05.02
There were a lot of bugfixes (not related to to you problem)
Next: if you see "read_params_file" in error message, it means (for 98%
;-)) your command for running tesseract is wrong. And you post everything
else but not how yo
What do you mean with "i haven't find the way to use it by dll"?
"is this possible?" yes it is. tesseract.exe use tesseract40.dll. So you
can use is at any other library.
Zdenko
pi 3. 5. 2019 o 9:16 Giuseppe Romano napísal(a):
> Hi,
> my name is Giuseppe, i have this problem. I use tesseract-o
Hi Lorenzo,
The link you shared was very helpful and your valuable suggestions helped
me a lot. Now the image detection seems to work for at least 90% of my
cases.
Just for information, I started with a top-hat transform followed by the
Sobel operator and then applied the Otsu thresholding. Th
Thanks for the answer but my tesseract version is the 3.05 and tesseract
40.dll in the
installation folder unfortunately not exist. Tesseract.exe use
libtesseract-3.dll and i haven't api
entry point, can you help me for the api declarations?
Thanks
Il giorno ven 3 mag 2019 alle ore 16:09 Zdenko Po
it is just older version, but basics are the same:
https://github.com/tesseract-ocr/tesseract/blob/3.05/api/baseapi.h
Or C-API:
https://github.com/tesseract-ocr/tesseract/blob/3.05/api/capi.h
Zdenko
pi 3. 5. 2019 o 16:57 Giuseppe Romano napísal(a):
> Thanks for the answer but my tesseract ver
Hi, Lorenzo,
Thank you very much for your reply. It really gives more clue about the
training.
All the best,
Tairen
On Friday, May 3, 2019 at 2:30:12 AM UTC-7, Lorenzo Blz wrote:
>
> See answer inline.
>
> Il giorno ven 3 mag 2019 alle ore 03:48 Tairen Chen > ha scritt
Thank you for your further explanation, Shree!!
On Friday, May 3, 2019 at 2:59:12 AM UTC-7, shree wrote:
>
> >There are three model sizes: best, normal and fast. Each of these can
> also be converted to an integer model.
>
> Only `best` can be converted to integer and in fact the LSTM models in
How did you add a blacklist?
On Monday, April 29, 2019 at 11:32:14 PM UTC-4, Jonathan wrote:
>
> If you know you won't have numbers, what worked for me is blacklisting
> numbers. Otherwise you will have to improve the image quality (like
> resizing to bigger size and sharping the edges)
>
> On M
The OCR gets more accurate with each blur... I've tried adding margins and
resizing and the results are the same. Anyone else have this issue?
--
You received this message because you are subscribed to the Google Groups
"tesseract-ocr" group.
To unsubscribe from this group and stop receiving e
Thank, i try to use this on my vb6 project.
Best regards
Giuseppe
Il giorno ven 3 mag 2019 alle ore 17:04 Zdenko Podobny
ha scritto:
> it is just older version, but basics are the same:
> https://github.com/tesseract-ocr/tesseract/blob/3.05/api/baseapi.h
>
> Or C-API:
> https://github.com/tesser
I perform fine tuning training by adding some extra data on the existing
model of ben.traindata, i named the "ds_10k.traindata", now how do i
perform the OCR task on this model ? In default, using the ben.traindata, i
just follow the command provided by tesseract, but now i wanna know do i
nee
17 matches
Mail list logo
