* scribo/toolchain/text_in_doc_preprocess.hh: New. * src/text_in_doc_preprocess.cc: Make use of this new toolchain. --- scribo/ChangeLog | 8 + scribo/scribo/toolchain/text_in_doc_preprocess.hh | 152 +++++++++++++++++++++ scribo/src/text_in_doc_preprocess.cc | 39 ++---- 3 files changed, 170 insertions(+), 29 deletions(-) create mode 100644 scribo/scribo/toolchain/text_in_doc_preprocess.hh diff --git a/scribo/ChangeLog b/scribo/ChangeLog index 5fe30ab..45ebaaf 100644 --- a/scribo/ChangeLog +++ b/scribo/ChangeLog @@ -1,5 +1,13 @@ 2010-08-25 Guillaume Lazzara <z@lrde.epita.fr> + Add a preprocessing toolchain. + + * scribo/toolchain/text_in_doc_preprocess.hh: New. + + * src/text_in_doc_preprocess.cc: Make use of this new toolchain. + +2010-08-25 Guillaume Lazzara <z@lrde.epita.fr> + Fix namespace ambiguities. * scribo/binarization/sauvola_ms.hh, diff --git a/scribo/scribo/toolchain/text_in_doc_preprocess.hh b/scribo/scribo/toolchain/text_in_doc_preprocess.hh new file mode 100644 index 0000000..cff99f7 --- /dev/null +++ b/scribo/scribo/toolchain/text_in_doc_preprocess.hh @@ -0,0 +1,152 @@ +// Copyright (C) 2010 EPITA Research and Development Laboratory (LRDE) +// +// This file is part of Olena. +// +// Olena is free software: you can redistribute it and/or modify it under +// the terms of the GNU General Public License as published by the Free +// Software Foundation, version 2 of the License. +// +// Olena is distributed in the hope that it will be useful, +// but WITHOUT ANY WARRANTY; without even the implied warranty of +// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU +// General Public License for more details. +// +// You should have received a copy of the GNU General Public License +// along with Olena. If not, see <http://www.gnu.org/licenses/>. +// +// As a special exception, you may use this file as part of a free +// software project without restriction. Specifically, if other files +// instantiate templates or use macros or inline functions from this +// file, or you compile this file and link it with other files to produce +// an executable, this file does not by itself cause the resulting +// executable to be covered by the GNU General Public License. This +// exception does not however invalidate any other reasons why the +// executable file might be covered by the GNU General Public License. + +#ifndef SCRIBO_TOOLCHAIN_TEXT_IN_DOC_PREPROCESS_HH +# define SCRIBO_TOOLCHAIN_TEXT_IN_DOC_PREPROCESS_HH + +/// \file +/// +/// \brief Preprocess a document before looking for its content. + + +#include <mln/core/concept/image.hh> +#include <mln/logical/not.hh> + +#include <scribo/binarization/sauvola_ms.hh> + +#include <scribo/preprocessing/split_bg_fg.hh> +#include <scribo/preprocessing/deskew.hh> + + +namespace scribo +{ + + namespace toolchain + { + + using namespace mln; + + /*! \brief Preprocess a document before looking for its content. + + \param[in] input An image. + \param[in] enable_fg_bg Enable/Disable background removal. + + If \p enable_fg_bg is set to 'True' then a background removal is + performed. Its parameter lambda is automatically set according + to the input image size. + + */ + template <typename I> + mln_ch_value(I,bool) + text_in_doc_preprocess(const Image<I>& input, bool enable_fg_bg); + + /*! \brief Preprocess a document before looking for its content. + + \param[in] input An image. + \param[in] lambda Parameter to the background removal. + + If lambda is set to '0' no background removal is + performed. Otherwise, a background removal is performed with the + given \p lambda value. + + */ + template <typename I> + mln_ch_value(I,bool) + text_in_doc_preprocess(const Image<I>& input, unsigned lambda); + + +# ifndef MLN_INCLUDE_ONLY + + + template <typename I> + mln_ch_value(I,bool) + text_in_doc_preprocess(const Image<I>& input_, bool enable_fg_bg) + { + trace::entering("scribo::toolchain::text_in_doc_preprocess"); + + const I& input = exact(input_); + mln_precondition(input.is_valid()); + + unsigned lambda = 0; + if (enable_fg_bg) + lambda = 1.2 * (input.nrows() + input.ncols()); + + mln_ch_value(I,bool) output = text_in_doc_preprocess(input, lambda); + + trace::exiting("scribo::toolchain::text_in_doc_preprocess"); + return output; + } + + + template <typename I> + mln_ch_value(I,bool) + text_in_doc_preprocess(const Image<I>& input_, unsigned lambda) + { + trace::entering("scribo::toolchain::text_in_doc_preprocess"); + + const I& input = exact(input_); + mln_precondition(input.is_valid()); + + mln_concrete(I) input_rgb = input; + + // Extract foreground + if (lambda != 0) + { + std::cout << "Extracting foreground..." << std::endl; + input_rgb = preprocessing::split_bg_fg(input, lambda, 32).second(); + } + + // Convert to Gray level image. + mln_ch_value(I,value::int_u8) + input_gl = data::transform(input_rgb, + mln::fun::v2v::rgb_to_int_u<8>()); + + + // Deskewing + std::cout << "Deskew if needed..." << std::endl; + input_gl = preprocessing::deskew(input_gl); + + // Binarize foreground to use it in the processing chain. + std::cout << "Binarizing foreground..." << std::endl; + mln_ch_value(I,bool) + input_bin = scribo::binarization::sauvola_ms(input_gl, 101, 3); + + logical::not_inplace(input_bin); + + trace::exiting("scribo::toolchain::text_in_doc_preprocess"); + return input_bin; + } + + +# endif // ! MLN_INCLUDE_ONLY + + + } // end of namespace scribo::toolchain + +} // end of namespace scribo + + +#endif // SCRIBO_TOOLCHAIN_TEXT_IN_DOC_PREPROCESS_HH + diff --git a/scribo/src/text_in_doc_preprocess.cc b/scribo/src/text_in_doc_preprocess.cc index fdbb018..b731272 100644 --- a/scribo/src/text_in_doc_preprocess.cc +++ b/scribo/src/text_in_doc_preprocess.cc @@ -34,17 +34,9 @@ #include <mln/value/rgb8.hh> -#include <mln/logical/not.hh> - -#include <scribo/binarization/sauvola_ms.hh> - #include <scribo/debug/usage.hh> -#include <scribo/preprocessing/split_bg_fg.hh> -#include <scribo/preprocessing/deskew.hh> - - -#include <mln/io/pgm/all.hh> +#include <scribo/toolchain/text_in_doc_preprocess.hh> const char *args_desc[][2] = @@ -74,28 +66,17 @@ int main(int argc, char* argv[]) unsigned lambda; if (argc == 5) lambda = atoi(argv[4]); - else - lambda = 1.2 * (input_rgb.nrows() + input_rgb.ncols()); - - // Extract foreground - if (argc >= 4 && atoi(argv[3]) == 1) - { - std::cout << "Extracting foreground..." << std::endl; - input_rgb = preprocessing::split_bg_fg(input_rgb, lambda, 32).second(); - } - // Convert to Gray level image. - image2d<value::int_u8> - input_gl = data::transform(input_rgb, mln::fun::v2v::rgb_to_int_u<8>()); + bool remove_bg = false; + if (argc >= 4) + remove_bg = (atoi(argv[3]) == 1); + image2d<bool> output; - // Deskewing - std::cout << "Deskew if needed..." << std::endl; - input_gl = preprocessing::deskew(input_gl); - - // Binarize foreground to use it in the processing chain. - std::cout << "Binarizing foreground..." << std::endl; - image2d<bool> input_bin = scribo::binarization::sauvola_ms(input_gl, 101, 3); + if (argc == 5 && remove_bg) + output = toolchain::text_in_doc_preprocess(input_rgb, lambda); + else + output = toolchain::text_in_doc_preprocess(input_rgb, remove_bg); - mln::io::pbm::save(input_bin, argv[2]); + mln::io::pbm::save(output, argv[2]); } -- 1.5.6.5