>
> Please see https://github.com/tesseract-ocr/tesseract/issues/837
>
This discussion is better held there.
--
You received this message because you are subscribed to the Google Groups
"tesseract-ocr" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to
Oops, I forgot the attachment. Here it is :-)
I believe it will help you further to decide, but what it CAN do:
- find whitelines
- map a zone to a certainn character probability
- train itself.
it does NOT decide whether it is a certain character or not, this needs to
be decided on the host, not
@zdenko This discussion maybe better suited for tesseract-dev forum or do
you want to track it as a issue on github?
ShreeDevi
भजन - कीर्तन - आरती @ http://bhajans.ramparivar.com
On Sat, Apr 28, 2018 at 1:19 PM, Janpieter Sollie
Would it be a problem for you if I rewrite the opencl engine completely,
and you people provide me help to link the tesseract kernel -> opencl
engine parts?
in attachment, I already have a list of features I'd like to port to
openCL. As this uses the GPU in a heavy way, I will implement
if I'm right, a neural net is about the engine parts, not the image
characterisation rendering method, am I right? because I see many
presentations, and most of them talk about the history of tesseract, but
that's not what I need
2018-04-27 14:27 GMT+00:00 ShreeDevi Kumar :
Please see
https://github.com/tesseract-ocr/tesseract/wiki/4.0-with-LSTM
For info about neural nets used by tesseract
On Fri 27 Apr, 2018, 7:48 PM Janpieter Sollie,
wrote:
> I had a quick thought about what you could offload to opencl. I will need
> some help from
I had a quick thought about what you could offload to opencl. I will need
some help from you people (I am a C programmer, not C++, at least not
experienced) to do the host code, but this algorithm is perfectly
optimizeable in openCL.
the way I'd do it:
prerequirements:
- you can define 65k
Okay, I took a quick look at the opencl code:
- you are only invoking 255 (16x16) OpenCL kernels at the same time (max).
This is not enough, not even for a budget GPU.
- the code has some points to discuss (eg: why does it keep the kernel and
the tesseract dat file in the local directory and
a thing which I could do quite easily is search for characters with a
certain offset. If you can give me 2^16 offsets
to analyze, It may be an advantage to do it via OpenCL. But tdoes the
network work this way?
2018-04-27 10:56 GMT+02:00 Janpieter Sollie :
> I'd be
I'd be glad to help. using tesseract 4, I am able to perform a 90%
accuracy on OpenCL. I do not have any experience with neural networks (i'm
just a high-school (no college educated IT-support guy with some knowledge
about OpenCL), so can you recommend me some documentation to understand the
If you have experience your help will be warmly welcomed.
OpenCL is not maintained and it is on good way to be removed if
maintainer/contributor will not be found.
Anyway it is not used extensively, so there is a place for improvement,
Zdenko
pi 27. 4. 2018 o 10:21 Janpieter Sollie
Hello everyone,
I have a question about the openCL selection procedure of tesseract:
my output:
[DS] Profile read from file (tesseract_opencl_profile_devices.dat).
[DS] Device[1] 1:Fiji score is 0.202927
[DS] Device[2] 1:Ellesmere score is 1.468799
[DS] Device[3] 1:Ellesmere score is 1.468799
12 matches
Mail list logo