psubhashish / parichit Goto Github PK

I want to make pre.tessdata but i came across with 2 errors 
1-when i type mftraining -F font_properties -U unicharset command

C:\Program Files (x86)\Tesseract-OCR>mftraining -F font_properties -U unicharset
 -O per.unicharset per.arial.exp0.tr
Failed to load font_properties
Reading per.arial.exp0.tr ...
arial has no defined properties.
!"Missing font_properties entry is a fatal error!":Error:Assert failed:in file .
.\training\mftraining.cpp, line 281


2-when I want to make  .tessdata file it shows 

C:\Program Files (x86)\Tesseract-OCR>combine_tessdata per.
Combining tessdata files
Error opening unicharset file
Error combining tessdata files into per.traineddata

I attached my box file and tif file would you please tell me what shoul I do?

What steps will reproduce the problem?
1.
2.
3.

What is the expected output? What do you see instead?


What version of the product are you using? On what operating system?


Please provide any additional information below.

What steps will reproduce the problem?
1. as the  parichit  ZIP did not have the tesseract  executable  and  engine  I 
tried  downloading the  latest version  and  copy the  executable to parichit 
folder  and  it  did  not  work  when  the  whole  zip file  was   extracted to 
the  tesseract   folder inside  parichit   it  gives  an  error  not  
accessable  for  some  files .

please  help  with  a step  by  step  installation  advise .. thanks .



2.
3.

What is the expected output? What do you see instead?


What version of the product are you using? On what operating system?


Please provide any additional information below.

What steps will reproduce the problem?
1. go to http://code.google.com/p/parichit/source/browse/
2. select trunk

The svn repository is empty. It says no files in the selected directory. Select 
a subdirectory, if one exists, or another directory from the tree on the left.

Please give a link so that the source code can be downloaded. I wanted to study 
how segmentation is done for Shirorekha languages.

Thanks and Reagrds
Puneet Goyal

What steps will reproduce the problem?
1. Using Tesseract tools for android
2. using the phone camera to take image
3. using the telugu trained data file.

What is the expected output? What do you see instead?
 recognised characters with garbage characters. -finally the output with errors.

What version of the product are you using? On what operating system?
developing on android platform,gingerbread.with ubuntu 12.04 as OS.

Please provide any additional information below.
thanks parichit for a trained data file.But the issue is that garbage random 
letters are being predicted at the output.


క్కమలక్ట క క క 

What steps will reproduce the problem?
1. take some image of hindi characters: yogini.tif
2. copy hin.traineddata in tessdata folder.
3. apply command: tesseract yogini.tif myresult -i hin

What is the expected output? What do you see instead?

ugly output
no interword spaces

What version of the product are you using? On what operating system?
tesseract 3.01
ubuntu 10.04

Please provide any additional information below.
please make clear how we can consider interword spaces into account plzzz

What steps will reproduce the problem?
1.Downloaded the Parichit_WIN_LIN.tar.gz 
2.
3.

What is the expected output? What do you see instead?

How To Install

What version of the product are you using? On what operating system?



Please provide any additional information below.

What steps will reproduce the problem?
1. Compile and install tesseract on Ubuntu 10.04 64 bit
2. Take attached image and using the tel.traineddata try to OCR

What is the expected output? What do you see instead?
Should give an ocred text output (like it does on Windows). Or like the 
tel.traineddata that I have does on Ubuntu.
But I get the following error instead.

$ tesseract papk12.tif papk12_par -l tel
actual_tessdata_num_entries_ <= TESSDATA_NUM_ENTRIES:Error:Assert failed:in 
file tessdatamanager.cpp, line 55
Segmentation fault


What version of the product are you using? On what operating system?
tesseract-3.00 compiled using gcc on Ubuntu 10.04
tel.traineddata provided here. 

Notes: 
The parichit tel.traineddata works fine on Windows 7 (32 bit compilation)
The tel.traineddata that I have works fine on Ubuntu 10.04 (64 bit)

Please provide any additional information below.