OCR3：tesseract script

通过命令：tesseract -h 可查看 OCR操作脚本参数：

• 

其中参数说明：

• –-oem：指定使用的算法，0：代表老的算法；1：代表LSTM算法；2：代表两者的结合；3：代表系统自己选择。
• –-psm：指定页面切分模式。默认是3，也就是自动的页面切分，但是不进行方向(Orientation)和文字(script，其实并不等同于文字，比如俄文和乌克兰文都使用相同的script，中文和日文的script也有重合的部分)的检测。如果我们要识别的是单行的文字，我可以指定7。我们这里已经知道文字是中文，并且方向是horizontal(从左往右再从上往下的写法，古代中国是从上往下从右往左），因此使用默认的3就可以了。

--psm:

• 

combine_tessdata

---

• -e：通过-e 指令，可以从一个已经合并了的traineddata文件中提取独立的组件。如：combine_tessdata -e tessdata/eng.traineddata \  /home/$USER/temp/eng.config /home/$USER/temp/eng.unicharset
• -o：通过 -o 指令，可以覆盖一个给定的traineddata文件中的对应组件。如：combine_tessdata -o tessdata/eng.traineddata \ /home/$USER/temp/eng.config /home/$USER/temp/eng.unicharambigs
• -u：通过 -u 指令，可以将所有组件解压到指定路径。如：combine_tessdata -u tessdata/eng.traineddata /home/$USER/temp/eng.

code

---

NAME
       combine_tessdata - combine/extract/overwrite/list/compact Tesseract data
        # 用于合并/提取/覆盖/list(-d)/压缩 tesseract data
SYNOPSIS
       combine_tessdata [OPTION] FILE...

DESCRIPTION
       combine_tessdata(1) is the main program to
       combine/extract/overwrite/list/compact tessdata components in
       [lang].traineddata files.
       # combine_tessdata 是主要的程序，用来合并/提取/覆盖/list/压缩 [lang].traineddata files 中的tessdata组件。

       To combine all the individual tessdata components (unicharset, DAWGs,
       classifier templates, ambiguities, language configs) located at, say,
       /home/$USER/temp/eng.* run:

           combine_tessdata /home/$USER/temp/eng.

       The result will be a combined tessdata file
       /home/$USER/temp/eng.traineddata
       # 将所有独立的tessdat组件合并在一起

       Specify option -e if you would like to extract individual components
       from a combined traineddata file. For example, to extract language
       config file and the unicharset from tessdata/eng.traineddata run:

           combine_tessdata -e tessdata/eng.traineddata \
             /home/$USER/temp/eng.config /home/$USER/temp/eng.unicharset

       The desired config file and unicharset will be written to
       /home/$USER/temp/eng.config /home/$USER/temp/eng.unicharset
       # 通过-e 指令，可以从一个已经合并了的traineddata文件中提取独立的组件。

       Specify option -o to overwrite individual components of the given
       [lang].traineddata file. For example, to overwrite language config and
       unichar ambiguities files in tessdata/eng.traineddata use:

           combine_tessdata -o tessdata/eng.traineddata \
             /home/$USER/temp/eng.config /home/$USER/temp/eng.unicharambigs

       As a result, tessdata/eng.traineddata will contain the new language
       config and unichar ambigs, plus all the original DAWGs, classifier
       templates, etc.
       # 通过 -o 指令，可以覆盖一个给定的traineddata文件中的对应组件。
       # 

       Note: the file names of the files to extract to and to overwrite from
       should have the appropriate file suffixes (extensions) indicating their
       tessdata component type (.unicharset for the unicharset, .unicharambigs
       for unichar ambigs, etc). See k*FileSuffix variable in
       ccutil/tessdatamanager.h.
       # 要提取和覆盖的文件的文件名应具有对应文件相同的后缀名，以表明其tessdata组件的类型。

       Specify option -u to unpack all the components to the specified path:
           combine_tessdata -u tessdata/eng.traineddata /home/$USER/temp/eng.

       This will create /home/$USER/temp/eng.* files with individual tessdata
       components from tessdata/eng.traineddata.
       # 通过 -u 指令，可以将所有组件解压到指定路径

OPTIONS
       -c .traineddata FILE...: Compacts the LSTM component in the
       .traineddata file to int.

       -d .traineddata FILE...: Lists directory of components from the
       .traineddata file.

       -e .traineddata FILE...: Extracts the specified components from the
       .traineddata file

       -o .traineddata FILE...: Overwrites the specified components of the
       .traineddata file with those provided on the command line.

       -u .traineddata PATHPREFIX Unpacks the .traineddata using the provided
       prefix.

CAVEATS
       Prefix refers to the full file prefix, including period (.)
       # 注意点 指令中的前缀要包含‘.’
COMPONENTS
#组件
       The components in a Tesseract lang.traineddata file as of Tesseract 4.0
       are briefly described below; For more information on many of these
       files, see
       https://github.com/tesseract-ocr/tesseract/wiki/TrainingTesseract and
       https://github.com/tesseract-ocr/tesseract/wiki/TrainingTesseract-4.00

       lang.config
           (Optional) Language-specific overrides to default config variables.
           For 4.0 traineddata files, lang.config provides control parameters
           which can affect layout analysis, and sub-languages.
           # 根据语言特定，用来覆盖默认的配置变量。
           # 对于4.0的traineddata文件来说，config文件提供影响布局分析（不知道跟文字分割算法有关）和子语言的控制参数

       lang.unicharset
           (Required - 3.0x legacy tesseract) The list of symbols that
           Tesseract recognizes, with properties. See unicharset(5).
           # 3.0 必需的
           # tesseract识别的符号列表，包含属性.

       lang.unicharambigs
           (Optional - 3.0x legacy tesseract) This file contains information
           on pairs of recognized symbols which are often confused. For
           example, rn and m.
           # 3.0 可选的
           # 这个文件包含经常容易混淆的符号对的信息，例如‘rn和m’
           # （如果识别中文的话，应该可以用来处理一些形似字，比如日和曰）

       lang.inttemp
           (Required - 3.0x legacy tesseract) Character shape templates for
           each unichar. Produced by mftraining(1).
           # 3.0 必需
           # 每个字符的形状模板
           # 通过模仿mftraining创建

       lang.pffmtable
           (Required - 3.0x legacy tesseract) The number of features expected
           for each unichar. Produced by mftraining(1) from .tr files.
           # 3.0 必需
           # 每个字符的期望特征数量
           # 由 mftraining 通过 .tr文件产生

       lang.normproto
           (Required - 3.0x legacy tesseract) Character normalization
           prototypes generated by cntraining(1) from .tr files.
           # 3.0 必需
           # 字符的归一化原型，由cntraining通过.tr文件生成

       lang.punc-dawg
           (Optional - 3.0x legacy tesseract) A dawg made from punctuation
           patterns found around words. The "word" part is replaced by a
           single space.
           # 3.0 可选的
           # 一个由字符周围标点符号构建的dawg
           # word部分由一个单独的空格替代

       lang.word-dawg
           (Optional - 3.0x legacy tesseract) A dawg made from dictionary
           words from the language.
           # 3.0 可选
           # 一个由字典单词构建的dawg

       lang.number-dawg
           (Optional - 3.0x legacy tesseract) A dawg made from tokens which
           originally contained digits. Each digit is replaced by a space
           character.
           # 3.0 可选
           # 一个由符号构建的dawg，最初包含数字，每一个数字被一个空格字符代替？？？不是很理解

      lang.freq-dawg
           (Optional - 3.0x legacy tesseract) A dawg made from the most
           frequent words which would have gone into word-dawg.
           # 3.0 可选
           # 一个由最常用单词构建的dawg，这些单词将会进入word-dwag

       lang.fixed-length-dawgs
           (Optional - 3.0x legacy tesseract) Several dawgs of different fixed
           lengths — useful for languages like Chinese.
           # 3.0 可选
           # 混合长度dawgs
           # 对类似于中文的语言有用
           # 

       lang.shapetable
           (Optional - 3.0x legacy tesseract) When present, a shapetable is an
           extra layer between the character classifier and the word
           recognizer that allows the character classifier to return a
           collection of unichar ids and fonts instead of a single unichar-id
           and font.
           # 3.0 可选
           # 如果存在，shapetable是字符分类器和单词识别器之间的额外层，允许字符分类器返回unichar ID和字体的集合，而不是单个unichar-id和字体。
           # （应该是指用来应对多字符识别的，应该能够提高准确率）

       lang.bigram-dawg
           (Optional - 3.0x legacy tesseract) A dawg of word bigrams where the
           words are separated by a space and each digit is replaced by a ?.
           # 一个由双字母组构成的dawg
           # bigram？？二元语法
           [wiki bigram](https://en.wikipedia.org/wiki/N-gram)

       lang.unambig-dawg
           (Optional - 3.0x legacy tesseract) .

       lang.params-model
           (Optional - 3.0x legacy tesseract) .

       lang.lstm
           (Required - 4.0 LSTM) Neural net trained recognition model
           generated by lstmtraining.
           # 4.0 必需
           # 由lstmtraining生成的神经网络识别模型

       lang.lstm-punc-dawg
           (Optional - 4.0 LSTM) A dawg made from punctuation patterns found
           around words. The "word" part is replaced by a single space. Uses
           lang.lstm-unicharset.
           # 4.0 可选
           # 由单词周边的标点符号构造的dawg，需要用到lang.lstm-unicharset

       lang.lstm-word-dawg
           (Optional - 4.0 LSTM) A dawg made from dictionary words from the
           language. Uses lang.lstm-unicharset.
           # 4.0 可选
           # 由指定的语言的字典单词构造的dawg，需要用到lang.lstm-unicharset

       lang.lstm-number-dawg
           (Optional - 4.0 LSTM) A dawg made from tokens which originally
           contained digits. Each digit is replaced by a space character. Uses
           lang.lstm-unicharset.
           # 4.0可选
           # 一个由最初包含数字的符号集构造的dawg
           # Each digit is replaced by a space character.这句话还是不是很理解，直译的话就是每个数字都由一个空格字符代替，
           我想或者是不是可以理解为每个数字都由一个空格字符所占用的位置代表？？
           # 需要用到lang.lstm-unicharset

       lang.lstm-unicharset
           (Required - 4.0 LSTM) The unicode character set that Tesseract
           recognizes, with properties. Same unicharset must be used to train
           the LSTM and build the lstm-*-dawgs files.
           # 4.0 必需
           # 一个Tesseract可以识别的包含属性的unicode字符集。
           # 相同的单字符集必须用来被训练LSTM，并且构造the lstm-*-dawgs files
       lang.lstm-recoder
           (Required - 4.0 LSTM) Unicharcompress, aka the recoder, which maps
           the unicharset further to the codes actually used by the neural
           network recognizer. This is created as part of the starter
           traineddata by combine_lang_model.
           # 4.0 必需
           # Unicharcompress又名the recoder （单字符压缩？又名编码器？）
           # 将单字符集合进一步映射到神经网络识别器实际使用到的代码上
           # 这个lang.lang_recoder由combine_lang_model创建的starter traineddata的一部分
           # （这个lang.lang_recoder可以通过combine_tessdata从traineddata中提取出来）

       lang.version
           (Optional) Version string for the traineddata file. First appeared
           in version 4.0 of Tesseract. Old version of traineddata files will
           report Version string:Pre-4.0.0. 4.0 version of traineddata files
           may include the network spec used for LSTM training as part of
           version string.
           # 4.0 可选
           # 为the traineddata file.创建的版本字符串
           #

HISTORY
       combine_tessdata(1) first appeared in version 3.00 of Tesseract

SEE ALSO
       tesseract(1), wordlist2dawg(1), cntraining(1), mftraining(1),
       unicharset(5), unicharambigs(5)

COPYING
       Copyright (C) 2009, Google Inc. Licensed under the Apache License,
       Version 2.0

AUTHOR
       The Tesseract OCR engine was written by Ray Smith and his research
       groups at Hewlett Packard (1985-1995) and Google (2006-present).

 参考资料

---

• https://fancyerii.github.io/2019/03/12/3_tesseract/
• https://blog.csdn.net/qq_32674197/article/details/81983444