Read PDF With OCR (Tesseract OCR)

Help Studio
, ,
1

Hi everyone,
I got a problem, which is when I read pdf file using tesseract OCR and get number but that’s not same with on pdf’s one.
Is there any solutions?

Regards,
Temuka

2

Specially doesn’t understand “8” or “9”

3

Try UiPath document ocr so that you may get the required value

1 Like
4

Hey,
Do I only have to write str variable on text from properties?

5

I think it’s not working.

6

Can you please share the sample data

7

Sorry, I can’t upload some reason.
And I already solved it.
Thank you!

8

@Temuulen_Buyangerel

Please change the scale of a teseract ocr in properties panel start from 1 and gradually increase by 0.5

9

I will try it
Thank you!
Regards

10

Have you tried Omni page OCR ?

11

Hey,
No I haven’t tried.
Still can’t get that exact number I want.

12

Hi @Temuulen_Buyangerel ,
Can you share your file ?
I will try read it
Regards,
LNV

13

Hey,
I can’t upload it, sorry.

14

Oh, I see
you can try read text PDF, because OCR sometime can not understand numeric
hope it help

1 Like
15

Hi,
You can see 2 way to read PDF
My file
image
Sample.pdf (805.1 KB)

  1. by OCR (Tesseract OCR)
    image
    image512×769 18.9 KB

    result
    image
    image876×225 3.56 KB

    It’s not good and too slow

2.by read PDF text

image
image533×724 18 KB

result
image
image710×125 3.07 KB

It’s better, faster
You can try
pdf _excel.xaml (15.7 KB)
fr.xlsx (11.8 KB)

Hope it help
Regards,
LNV

1 Like
16

Thank you!
It works.
Regards,
Temuulen

17

Sometimes, simple is better than complicated :smile:
Cheer

18

There were some texts not in English so, I was using Tesseract and traineddata of other language.
Now I used other read pdf activity to get numeric.
Thanks for solution!

1 Like
19

Yes, some pictographic languages like china, japan,… OCR is better
happy automation

20

This topic was automatically closed 3 days after the last reply. New replies are no longer allowed.