search

shadowkane / serak-tesseract-trainer

Automatically exported from code.google.com/p/serak-tesseract-trainer
0 stars 0 forks source link

Box File Not Found With The Image Folder #1

Open GoogleCodeExporter opened 8 years ago

GoogleCodeExporter commented 8 years ago 
What steps will reproduce the problem?
1. When adding new image .tiff file, the summary message appears
2. Choices are: Create New Box with help saying its based on default lang (eng)
3. Or Bootstrapping a new character set with help saying it will create a new 
character set for your lang.

What is the expected output? What do you see instead?
In video tutorial this screen doesnt appear, and using either option when 
Combining tessdata the file: 'normproto' is missing.
I would like to know how to generate this normproto file.
Couldnt find the answer online anywhere.

What version of the product are you using? On what operating system?
Serak trainer for tesseract 3.0X (the most recent availible)
Tessract 3.02.02 (installed with .exe)
Windows 8 pro 64-bit

Please provide any additional information below.
i attached 2 print screen of error messages.
1 is the error when adding the image.
2 is the erros when combining tessData.

Thanks ahead for any feedback.
Feel free to contact to my e-mail: jose.miguel.loureiro@gmail.com

Original issue reported on code.google.com by jose.mig...@gmail.com on 18 Apr 2013 at 2:43

Attachments:

GoogleCodeExporter commented 8 years ago 
first make sure the box file is created in your project folder and make sure 
the file ( *.box ) file is not empty use . if you need all the commands for 
tesseract  goto this link 

http://code.google.com/p/tesseract-ocr/wiki/TrainingTesseract3

but, i dont think your problem is from the trainer. -> first your have to make 
sure. the correct .box file has been created , because sometimes i get an empty 
.box files too and when you use those box files to create. 'normproto' it skips 
and u can combine without the normproto file.

so i recommend checking the box file using 'qt box creator ' or similar 3rp 
party application. then get back to the trainer and and import the tiff file.

Original comment by sirak2...@gmail.com on 30 Apr 2013 at 6:21

GoogleCodeExporter commented 8 years ago 
hi,

I am having same problem.
I am working on win7.

any solution for this..???

thanking you in advance.

Original comment by hiravsha...@gmail.com on 26 Jul 2013 at 11:25

GoogleCodeExporter commented 8 years ago 
I am also faing the same normproto file not found error. I tried creating a 
dummy blank normproto file and proceeded on step 4 to combine data.. but OCR 
recogntion ran into error with the so generated traindata.  
Steps I had followed on windows7.

1. Installed new fonts ttf. Seven Segment ttf file for 0-9 in seven segment 
display style.
2. Created a text file with 0-9 and decimal.
3. Installed JTessBoxEditor and created TIFF/BOX file from the above text file. 
Verified box created OK both in the tool and through notepad.
4. unzipped Tesseract 3.02 for win32.
5.  Using SerakTrainer created a .ser project, provided Tesseract path, set the 
language and imported the image file and trained Tesseract.
6. Got message successfully trained.
7. All files generated in Traindata except "normproto"
8. On clicking 4 step - combine data, shows error file not found for normproto.

Original comment by mubeenkh...@gmail.com on 12 Sep 2014 at 5:17

GoogleCodeExporter commented 8 years ago 
Did anyone figure this out? Running into this normproto issue now, can't figure 
out what's wrong.

Original comment by joshomat...@gmail.com on 17 Nov 2014 at 6:46