Tesseract-OCR 训练过程 V3.02

软件：

jTessBoxEditor Version 0.9 (30 April 2013)

Tesseract-OCR win32 v3.02 with Leptonica

 

训练步骤：

 

1.使用jTessBoxEditor,tools->merge_tif，产生tif文件

2.产生box文件

tesseract.exe eng.arial.01.tif eng.arial.01 batch.nochop makebox

3.使用jTessBoxEditor打开，Insert或Delete，添加删除字符，并通过xywh调整对应的坐票

4.训练(如果遇到不可识别的字符，couldn t find a matching blob，尝试换位置或调坐标)

tesseract.exe eng.arial.01.tif eng.arial.01 nobatch box.train

5.字体预处理

unicharset_extractor.exe eng.arial.01.box

6.创建font_properties.txt，内容为：arial 0 0 0 0 0

7.字体处理

mftraining.exe -F font_properties.txt -U unicharset eng.arial.01.tr

8.cntraining.exe eng.arial.01.tr

9.把unicharset, inttemp, normproto, pffmtable这四个文件加上前缀“eng.arial.01.”

10.combine_tessdata.exe eng.arial.01.

 

显示：

Combining tessdata files

TessdataManager combined tesseract data files.

Offset for type 0 is -1

Offset for type 1 is 108

Offset for type 2 is -1

Offset for type 3 is 1660

Offset for type 4 is 327545

Offset for type 5 is 327781

Offset for type 6 is -1

Offset for type 7 is -1

Offset for type 8 is -1

Offset for type 9 is -1

Offset for type 10 is -1

Offset for type 11 is -1

Offset for type 12 is –1

 

必须确定的是第2、4、5、6行的数据不是-1，那么一个新的字典就算生成了。

 

11.此时目录下“eng.arial.01.traineddata”的文件拷贝到tesseract程序目录下的“tessdata”目录

12.

#tesseract.exe test.jpg result -l eng.arial.01

#tesseract.exe a.bmp result2 -l eng.arial.01

 

指定布局识别方式

tesseract.exe 42.png result2 -l eng.arial.01 -psm 7

 

 

布局参数描述：

 

-psm N

    Set Tesseract to only run a subset of layout analysis and assume a certain form of image. The options for N are:

 

    0 = Orientation and script detection (OSD) only.

    1 = Automatic page segmentation with OSD.

    2 = Automatic page segmentation, but no OSD, or OCR.

    3 = Fully automatic page segmentation, but no OSD. (Default)

    4 = Assume a single column of text of variable sizes.

    5 = Assume a single uniform block of vertically aligned text.

    6 = Assume a single uniform block of text.

    7 = Treat the image as a single text line.

    8 = Treat the image as a single word.

    9 = Treat the image as a single word in a circle.

    10 = Treat the image as a single character.