Tesseract-orc Merge Recognition Results

In the actual use of the TESSERACT-ORC recognition Library, the first production of the identification database is likely to be less than ideal recognition rate, need to slowly add later

This article shows how to combine multiple modified box files into a single recognition library.

First, you need a picture sample. tif file, location file. box, as long as you have these two files in, you can merge the dictionary

Suppose you already have the following sample picture and a modified box file:

Image.font.1.tif Image.font.1.box

Image.font.2.tif Image.font.2.box

Image.font.3.fit Image.font.3.box

1. tr file corresponding to Mr.

Tesseract image.font.1.tif image.font.1 Nobatch box.train

Tesseract image.font.2.tif image.font.2 Nobatch box.train

Tesseract image.font.3.tif image.font.3 Nobatch box.train

2. Extracting characters

Unicharset_extractor Image.font.1.box Image.font.2.box Image.font.3.box

3. Create a font signature file

Create a new Font_properties file (note that there is no suffix) to add all the font features of the box file

Font 0 0 0 0 0

4. Execute the following command

Mftraining-f font-u unicharset image.font.1.tr image.font.2.tr image.font.3.tr

5. Gather all. tr files

Cntraining image.font.1.tr image.font.2.tr image.font.3.tr

6. Renaming files

Rename the following file, add the font name in front, I use "CK" here

Unicharset

Inttemp

Normproto

Pfftable

Shapetable----Many tutorials have leaked this file, do not change this file to create a recognition library when the error.

After renaming, the file names are as follows

Ck.unicharset

Ck.inttemp

Ck.normproto

Ck.pfftable

Ck.shapetable----Many tutorials have leaked this file, do not change this file to create a recognition library when the error.

7. Merge all files to generate a large font file

Combine_tessdata CK.

Finish the call.

Tesseract-orc Merge Recognition Results