Use the latest code - it should be fixed.
Zdenko
po 6. 5. 2019 o 7:02 napísal(a):
>
> Hello I'm doing a simple command like this:
>
> tesseract thumb0546.jpg outputbase tsv
>
>
> The issue is that for one of the words, the letter 'a' it's giving me the
> full image size as the rect containing
Hello I'm doing a simple command like this:
tesseract thumb0546.jpg outputbase tsv
The issue is that for one of the words, the letter 'a' it's giving me the
full image size as the rect containing the word.
5 1 1 1 3 2 *0 0 640 360* 96 a
I'm using OS X. Here's the version info. Image and f
I am not sure what OS you use, but AFAIK ImageMagick should use internally
ghostscript.
After several testing (in other project) I found this command (windows
version, for other OS you need to find correct name of ghostscript
executable) for converting pdf to tiff:
gswin64c.exe -dBATCH -dTextAlpha
*the current tool am using is image magic I tried to convert the PDF to
Image using another tool, and it seems the result did come out correct*
On Sunday, May 5, 2019 at 4:19:15 PM UTC+2, shree wrote:
>
> Problem seems to be with with the jpg image that you are using.
>
> Get correct results when
Any recommendation on how to convert pdf to image to be used with
tesseract-ocr then ? any preferred tool to do that ?
On Sunday, May 5, 2019 at 4:19:15 PM UTC+2, shree wrote:
>
> Problem seems to be with with the jpg image that you are using.
>
> Get correct results when using the pdf file with
*same reuslt, find the output file at *
*https://drive.google.com/open?id=1XTwFeMmvQ4Mka-3_rXTaGVUJ9b4OwLta*
On Sunday, May 5, 2019 at 4:05:21 PM UTC+2, shree wrote:
>
> Try with max-iterations 400
>
> On Sun, May 5, 2019 at 7:33 PM fady taher > wrote:
>
>> *I used option --fontlist "Calibri"
Problem seems to be with with the jpg image that you are using.
Get correct results when using the pdf file with gimagereader.
https://www.illinoiscapacitor.com/pdf/generated/lytics_products_detail_11040.pdf
Frequency Multipliers:
50 HZ 120 HZ 400 HZ 1 KHZ 10 KHZ 100 KHZ
0.9 1 1 1.15 (1/125
Try with max-iterations 400
On Sun, May 5, 2019 at 7:33 PM fady taher wrote:
> *I used option --fontlist "Calibri" and --max_iterations 3600*
>
>
> On Sunday, May 5, 2019 at 4:02:05 PM UTC+2, shree wrote:
>>
>> Which font did you use? Hopefully it was similar to your image. How many
>> iterati
*I used option --fontlist "Calibri" and --max_iterations 3600*
On Sunday, May 5, 2019 at 4:02:05 PM UTC+2, shree wrote:
>
> Which font did you use? Hopefully it was similar to your image. How many
> iterations?
>
> On Sun, May 5, 2019 at 6:58 PM fady taher > wrote:
>
>> *I followed the instru
Which font did you use? Hopefully it was similar to your image. How many
iterations?
On Sun, May 5, 2019 at 6:58 PM fady taher wrote:
> *I followed the instructions*
> https://github.com/tesseract-ocr/tesseract/wiki/TrainingTesseract-4.00.md#fine-tuning-for-impact
> ,
> *I added (S) for about 1
Share an image for testing.
How did you try to finetune?
On Sunday, May 5, 2019 at 5:40:39 PM UTC+5:30, fady taher wrote:
>
> *I do have numbers but this character "S" is pretty obvious, yet I think
> it keeps recognizing it with wrong value "5" due to the parentheses"(" and
> ")"*
>
> On Tues
*I do have numbers but this character "S" is pretty obvious, yet I think it
keeps recognizing it with wrong value "5" due to the parentheses"(" and ")"*
On Tuesday, April 30, 2019 at 5:32:14 AM UTC+2, Jonathan Muller wrote:
>
> If you know you won't have numbers, what worked for me is blacklistin
Thanks, I found it under /usr/local/cuda-10.0/targets/x86_64-linux/lib/ and
added it "export LDFLAGS=-L/usr/local/cuda-10.0/targets/x86_64-linux/lib/"
worked
On Wednesday, 1 May 2019 17:01:32 UTC+3, JB Data31 wrote:
>
> If my memories are goods you have to combine 2 options :
> -L
> -l <"the na
13 matches
Mail list logo
