gnu/packages/ocr.scm

   1 ;;; GNU Guix --- Functional package management for GNU
   2 ;;; Copyright © 2013 Ludovic Courtès <ludo@gnu.org>
   3 ;;; Copyright © 2016, 2020 Efraim Flashner <efraim@flashner.co.il>
   4 ;;; Copyright © 2019 Tobias Geerinckx-Rice <me@tobias.gr>
   5 ;;; Copyright © 2019 Alex Vong <alexvong1995@gmail.com>
   6 ;;;
   7 ;;; This file is part of GNU Guix.
   8 ;;;
   9 ;;; GNU Guix is free software; you can redistribute it and/or modify it
  10 ;;; under the terms of the GNU General Public License as published by
  11 ;;; the Free Software Foundation; either version 3 of the License, or (at
  12 ;;; your option) any later version.
  13 ;;;
  14 ;;; GNU Guix is distributed in the hope that it will be useful, but
  15 ;;; WITHOUT ANY WARRANTY; without even the implied warranty of
  16 ;;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  17 ;;; GNU General Public License for more details.
  18 ;;;
  19 ;;; You should have received a copy of the GNU General Public License
  20 ;;; along with GNU Guix.  If not, see <http://www.gnu.org/licenses/>.
  21
  22 (define-module (gnu packages ocr)
  23   #:use-module ((guix licenses) #:prefix license:)
  24   #:use-module (guix packages)
  25   #:use-module (guix download)
  26   #:use-module (guix git-download)
  27   #:use-module (guix build-system gnu)
  28   #:use-module (guix build-system python)
  29   #:use-module (gnu packages compression)
  30   #:use-module (gnu packages python)
  31   #:use-module (gnu packages image))
  32
  33 (define-public ocrad
  34   (package
  35     (name "ocrad")
  36     (version "0.27")
  37     (source (origin
  38              (method url-fetch)
  39              (uri (string-append "mirror://gnu/ocrad/ocrad-"
  40                                  version ".tar.lz"))
  41              (sha256
  42               (base32
  43                "0divffvcaim89g4pvqs8kslbcxi475bcl3b4ynphf284k9zfdgx9"))))
  44     (build-system gnu-build-system)
  45     (native-inputs `(("lzip" ,lzip)))
  46     (home-page "https://www.gnu.org/software/ocrad/")
  47     (synopsis "Optical character recognition based on feature extraction")
  48     (description
  49      "GNU Ocrad is an optical character recognition program based on a
  50 feature extraction method.  It can read images in PBM, PGM or PPM formats and
  51 it produces text in 8-bit or UTF-8 formats.")
  52     (license license:gpl3+)))
  53
  54 (define-public tesseract-ocr
  55   (package
  56     (name "tesseract-ocr")
  57     (version "3.04.01")
  58     (source
  59      (origin
  60        (method git-fetch)
  61        (uri (git-reference
  62               (url "https://github.com/tesseract-ocr/tesseract")
  63               (commit version)))
  64        (file-name (git-file-name name version))
  65        (sha256
  66         (base32 "0h1x4z1h86n2gwknd0wck6gykkp99bmm02lg4a47a698g4az6ybv"))))
  67     (build-system gnu-build-system)
  68     (inputs
  69      `(("leptonica" ,leptonica)))
  70     (arguments
  71      '(#:configure-flags
  72        (let ((leptonica (assoc-ref %build-inputs "leptonica")))
  73          (list (string-append "LIBLEPT_HEADERSDIR=" leptonica "/include")))))
  74     (home-page "https://github.com/tesseract-ocr/tesseract")
  75     (synopsis "Optical character recognition engine")
  76     (description
  77      "Tesseract is an optical character recognition (OCR) engine with very
  78 high accuracy.  It supports many languages, output text formatting, hOCR
  79 positional information and page layout analysis.  Several image formats are
  80 supported through the Leptonica library.  It can also detect whether text is
  81 monospaced or proportional.")
  82     (license license:asl2.0)))
  83
  84 (define-public zinnia
  85   (let* ((commit "581faa8f6f15e4a7b21964be3a5ec36265c80e5b")
  86          (revision "1")
  87          ;; version copied from 'configure.in'
  88          (version (git-version "0.07" revision commit)))
  89     (package
  90       (name "zinnia")
  91       (version version)
  92       (source
  93        (origin
  94          (method git-fetch)
  95          (uri (git-reference
  96                (url "https://github.com/taku910/zinnia")
  97                (commit commit)))
  98          (sha256
  99           (base32
 100            "1izjy5qw6swg0rs2ym2i72zndb90mwrfbd1iv8xbpwckbm4899lg"))
 101          (file-name (git-file-name name version))
 102          (modules '((guix build utils)
 103                     (ice-9 ftw)
 104                     (srfi srfi-26)))
 105          (snippet ; remove unnecessary files with potentially different license
 106           '(begin
 107              (for-each delete-file-recursively
 108                        (scandir "."
 109                                 (negate (cut member <> '("zinnia"
 110                                                          "." "..")))))
 111              #t))))
 112       (build-system gnu-build-system)
 113       (arguments
 114        `(#:phases
 115          (modify-phases %standard-phases
 116            (replace 'bootstrap
 117              (lambda _
 118                (chdir "zinnia")
 119                (for-each make-file-writable
 120                          '("config.log" "config.status"))
 121                #t)))))
 122       (home-page "https://taku910.github.io/zinnia/")
 123       (synopsis "Online hand recognition system with machine learning")
 124       (description
 125        "Zinnia is a simple, customizable and portable online hand recognition
 126 system based on Support Vector Machines.  Zinnia simply receives user pen
 127 strokes as a sequence of coordinate data and outputs n-best characters sorted
 128 by SVM confidence.  To keep portability, Zinnia doesn't have any rendering
 129 functionality.  In addition to recognition, Zinnia provides training module
 130 that allows us to create any hand-written recognition systems with low-cost.")
 131       (license (list license:bsd-3 ; all files except...
 132                      (license:non-copyleft ; some autotools related files
 133                       "file://zinnia/aclocal.m4")
 134                      license:x11 ; 'install-sh'
 135                      license:public-domain))))) ; 'install-sh'
 136
 137 ;;; python 2 bindings, license under the same terms as zinnia
 138 (define-public python2-zinnia
 139   (package
 140     (inherit zinnia)
 141     (name "python2-zinnia")
 142     (build-system python-build-system)
 143     (arguments
 144      `(#:python ,python-2 ; CObject API is used, it was removed in Python 3.2
 145        #:tests? #f ; avoid circular dependency on tegaki-zinnia-japanese
 146        #:phases
 147        (modify-phases %standard-phases
 148          (add-after 'unpack 'chdir
 149            (lambda _
 150              (chdir "zinnia/python")
 151              #t)))))
 152     (inputs
 153      `(("zinnia" ,zinnia)))))