| 1 | -------------------------------------------------
|
|---|
| 2 | CONTENTS
|
|---|
| 3 | -------------------------------------------------
|
|---|
| 4 | In this file:
|
|---|
| 5 |
|
|---|
| 6 | A. COMPILING TESSERACT GS2-EXTENSION
|
|---|
| 7 | & CREATING THE CUT-DOWN BINARY-ONLY TARBALL
|
|---|
| 8 |
|
|---|
| 9 | B. GETTING TIKA AND TESSERACT TO OCR A PDF
|
|---|
| 10 |
|
|---|
| 11 |
|
|---|
| 12 | -------------------------------------------------
|
|---|
| 13 | A. COMPILING TESSERACT GS2-EXTENSION
|
|---|
| 14 | & CREATING THE CUT-DOWN BINARY-ONLY TARBALL
|
|---|
| 15 | -------------------------------------------------
|
|---|
| 16 |
|
|---|
| 17 | To compile the Tesseract gs2-extension and then create the "binary" tarball needed to run
|
|---|
| 18 | Tesseract, we follow an equivalent version of the instructions for the imagemagick gs2-extension
|
|---|
| 19 | at http://trac.greenstone.org/browser/gs2-extensions/imagemagick/trunk/README
|
|---|
| 20 |
|
|---|
| 21 |
|
|---|
| 22 | 1. Find a location on your machine
|
|---|
| 23 |
|
|---|
| 24 |
|
|---|
| 25 | 2. Check out the tesseract extension from gs2-extensions
|
|---|
| 26 | svn co http://trac.greenstone.org/browser/gs2-extensions/tesseract/trunk tesseract
|
|---|
| 27 |
|
|---|
| 28 |
|
|---|
| 29 | 3. Compile it all up (tesseract and dependencies):
|
|---|
| 30 | cd tesseract
|
|---|
| 31 | ./CASCADE-MAKE.sh
|
|---|
| 32 |
|
|---|
| 33 |
|
|---|
| 34 | 4. Open a fresh terminal and check that the tesseract now installed in src/linux/bin works:
|
|---|
| 35 |
|
|---|
| 36 | cd src
|
|---|
| 37 | source ./setup.bash
|
|---|
| 38 |
|
|---|
| 39 | This should have set up env vars like GEXTTESS, GEXTTESS_INSTALLED, and TESSDATA_PREFIX
|
|---|
| 40 | which Tesseract needs to have set
|
|---|
| 41 |
|
|---|
| 42 | tesseract --list-langs
|
|---|
| 43 | tesseract sample.tif out
|
|---|
| 44 |
|
|---|
| 45 | OCRs sample.tif and generates out.txt from it.
|
|---|
| 46 |
|
|---|
| 47 | cat out.txt
|
|---|
| 48 |
|
|---|
| 49 | If you run Tesseract with the hocr config file, you can get the OCR output in
|
|---|
| 50 | nicely formatted html more representative of the input structure:
|
|---|
| 51 |
|
|---|
| 52 | tesseract sample.tif hocrtest
|
|---|
| 53 |
|
|---|
| 54 | The OCR output in html format will be in hocrtest.hocr:
|
|---|
| 55 |
|
|---|
| 56 | cat hocrtest.hocr
|
|---|
| 57 |
|
|---|
| 58 |
|
|---|
| 59 | 5. If successful, create the cut down tesseract binary zip and tarball by running the following
|
|---|
| 60 | at the toplevel of the extension checkout:
|
|---|
| 61 |
|
|---|
| 62 | ./makedists.sh <linux-x64|linux>
|
|---|
| 63 |
|
|---|
| 64 |
|
|---|
| 65 | If manually creating the cut-down tesseract zip and tarball then:
|
|---|
| 66 | a. create a folder at the same level as src called tesseract
|
|---|
| 67 | cd src
|
|---|
| 68 | cd ..
|
|---|
| 69 | mkdir tesseract
|
|---|
| 70 |
|
|---|
| 71 | b. COPY the setup files and MOVE the installed folder (src/linux) into the new cut-down tesseract folder:
|
|---|
| 72 |
|
|---|
| 73 | cp src/setup.ba* tesseract/.
|
|---|
| 74 | mv src/linux tesseract/.
|
|---|
| 75 |
|
|---|
| 76 | c. COPY the TESSERACT-APACHE-LICENSE and LEPTONICA-LICENSE txt files (note it uses
|
|---|
| 77 | American spelling!) from src/packages into the cut-down tesseract/linux:
|
|---|
| 78 |
|
|---|
| 79 | cp src/packages/*LICENSE.txt tesseract/linux/.
|
|---|
| 80 |
|
|---|
| 81 | d. Copy the top-level GETTING-OCR-SUPPORT-FOR-MORE-LANGS.txt file into the cutdown tesseract:
|
|---|
| 82 | cp GETTING-OCR-SUPPORT-FOR-MORE-LANGS.txt tesseract/.
|
|---|
| 83 |
|
|---|
| 84 | e. REMOVE folder "man" from tesseract/linux:
|
|---|
| 85 | rm -rf tesseract/linux/man
|
|---|
| 86 |
|
|---|
| 87 | f. REMOVE everything EXCEPT the "tessdata" folder in tesseract/linux/share.
|
|---|
| 88 | (The other things in that location are either unnecessary or created by tesseract's dependencies).
|
|---|
| 89 |
|
|---|
| 90 |
|
|---|
| 91 | 6. Create a tarball of the cut down tesseract folder named tesseract-<os>-<arch>.tar.gz:
|
|---|
| 92 | tar -cvzf tesseract-linux-x64.tar.gz tesseract
|
|---|
| 93 |
|
|---|
| 94 |
|
|---|
| 95 | 7. (Add/SVN up and) commit that to svn:
|
|---|
| 96 | svn up
|
|---|
| 97 | svn add tesseract-linux-x64.tar.gz
|
|---|
| 98 | (or svn diff tesseract-linux-x64.tar.gz if there was an earlier version to confirm modified)
|
|---|
| 99 | svn commit -m "MESSAGE" tesseract-linux-x64.tar.gz
|
|---|
| 100 |
|
|---|
| 101 |
|
|---|
| 102 | -------------------------------------------------
|
|---|
| 103 | B. GETTING TIKA AND TESSERACT TO OCR A PDF
|
|---|
| 104 | -------------------------------------------------
|
|---|
| 105 | Tesseract does not OCR PDFs (https://github.com/tesseract-ocr/tesseract/issues/1476).
|
|---|
| 106 | Trying to do so, you'll see:
|
|---|
| 107 | tesseract pdf05-notext.pdf notext
|
|---|
| 108 | Tesseract Open Source OCR Engine v5.0.0-alpha-694-g6ee3 with Leptonica
|
|---|
| 109 | Error in pixReadStream: Pdf reading is not supported
|
|---|
| 110 | Error in pixRead: pix not read
|
|---|
| 111 | Error during processing.
|
|---|
| 112 |
|
|---|
| 113 | Tesseract can OCR the individual images constituting a page of the PDF, but to OCR PDFs
|
|---|
| 114 | with Tesseract, you need an additional tool to split PDFs into its pages and extract images
|
|---|
| 115 | from them, feed each page's image into Tesseract to get it OCR-ed and then create an html or
|
|---|
| 116 | txt file collating all the individual OCR-ed page content.
|
|---|
| 117 |
|
|---|
| 118 | Tika does this.
|
|---|
| 119 |
|
|---|
| 120 | By default if Tika is on the environment and TESSDATA_PREFIX is set to the tessdata folder
|
|---|
| 121 | containing the language files, Tika is able to get Tesseract to OCR images out of the box.
|
|---|
| 122 | Yet not PDFs. Tika will output empty OCR for (x)html/txt, extracting only extractable text
|
|---|
| 123 | from PDFs and no OCR until the following is correct.
|
|---|
| 124 |
|
|---|
| 125 | To get Tika (app v1.24.x) and Tesseract (v5.0.0) set up to OCR PDFs, needed to do 2 more
|
|---|
| 126 | things:
|
|---|
| 127 | 1. Have to run the tika-app-*.jar with in --config=/path/to/a/tika-config.xml file
|
|---|
| 128 | configured correctly for the TesseractOCRParser and PDFParser
|
|---|
| 129 | 2. The <tika-config.xml> file passed to tika-app-*.jar should configure the "outputType"
|
|---|
| 130 | param's configuration of the TesseractOCRParser as follows:
|
|---|
| 131 | a. Set the "outputType" param to "txt" so Tesseract produces the OCR in .txt format, and
|
|---|
| 132 | Tesseract will produce .txt as OCR output which Tika will intercept and process,
|
|---|
| 133 | b. OR if you want the OCR generated by Tesseract to be in hocr (html ocr) format, then set
|
|---|
| 134 | the "outputType" param value to "hocr" AND ensure a config file also called hocr exists in
|
|---|
| 135 | $TESSDATA_PREFIX/configs containing the following (taken from
|
|---|
| 136 | https://github.com/tesseract-ocr/tessconfigs/blob/3decf1c8252ba6dbeef0bf908f4b0aab7f18d113/configs/hocr):
|
|---|
| 137 | tessedit_create_hocr 1
|
|---|
| 138 | hocr_font_info 0
|
|---|
| 139 |
|
|---|
| 140 |
|
|---|
| 141 |
|
|---|
| 142 | In order to have a tessdata/configs/hocr, needed to correct the Tesseract we were
|
|---|
| 143 | cascade-making to get it to put the "configs" subfolder inside the installed Tessereact's
|
|---|
| 144 | tessdata folder. The source version of tesseract has this folder, but it wasn't getting
|
|---|
| 145 | included in the built version despite us exporting TESSDATA_PREFIX before CASCADE-MAKing.
|
|---|
| 146 |
|
|---|
