Index: olena/ChangeLog from Giovanni Palma &lt;giovanni(a)lrde.epita.fr&gt; * tests/convol/tests/md5_convol_fast_gaussian_hh5: Correct test. * tests/convol/tests/md5_convol_fast_gaussian_hh4: Likewise. * tests/convol/tests/md5_convol_fast_gaussian_hh3: Likewise. * tests/convol/tests/md5_convol_fast_gaussian_hh2: Likewise. * tests/convol/tests/md5_convol_fast_gaussian_hh1: Likewise. * tests/convol/tests/md5_convol_fast_gaussian_hh0: Likewise. * oln/convol/fast_gaussian.hh: Add comments. * oln/convol/fast_gaussian.hxx: Likewise. Index: olena/oln/convol/fast_gaussian.hh --- olena/oln/convol/fast_gaussian.hh Fri, 19 Mar 2004 14:01:37 +0100 palma_g (oln/26_fast_gauss 1.6.1.6.1.7 600) +++ olena/oln/convol/fast_gaussian.hh Mon, 29 Mar 2004 15:22:05 +0200 palma_g (oln/26_fast_gauss 1.6.1.6.1.7 640) @@ -67,6 +67,8 @@ ** \arg sigma Value of sigma when computing the gaussian. ** \arg behavior Object to know how to work on borders. ** + ** \warning If sigma is big enough, the result may differ a + ** little bit with the -O3 flag of g++. */ template <class C, class B, class I, class BE> typename mute<I, typename convoutput<C, B, oln_value_type(I)>::ret>::ret @@ -90,6 +92,9 @@ ** \arg sigma Value of sigma when computing the gaussian. ** \arg behavior Object to know how to work on borders. ** + ** \warning If sigma is big enough, the result may differ a + ** little bit with the -O3 flag of g++. + ** */ template <class C, class B, class I, class BE> typename mute<I, typename convoutput<C, B, oln_value_type(I)>::ret>::ret @@ -113,6 +118,9 @@ ** \arg sigma Value of sigma when computing the gaussian. ** \arg behavior Object to know how to work on borders. ** + ** \warning If sigma is big enough, the result may differ a + ** little bit with the -O3 flag of g++. + ** */ template <class C, class B, class I, class BE> typename mute<I, typename convoutput<C, B, oln_value_type(I)>::ret>::ret @@ -159,6 +167,9 @@ ** => ** \image html oln_convol_fast_gaussian.png ** \image latex oln_convol_fast_gaussian.png + ** + ** \warning If sigma is big enough, the result may differ a + ** little bit with the -O3 flag of g++. */ template <class I, class BE> inline oln_concrete_type(I) @@ -191,7 +202,7 @@ ** ** im_type im1(oln::load(IMG_IN "lena256.pgm")); ** - ** oln::save(oln::convol::fast::gaussian(im1, 2, + ** oln::save(oln::convol::fast::gaussian_derivative(im1, 2, ** oln::value_behavior<oln_value_type_(im_type)>(0)), ** IMG_OUT "oln_convol_fast_gaussian_derivative.pbm"); ** return 0; @@ -204,7 +215,8 @@ ** \image html oln_convol_fast_gaussian_derivative.png ** \image latex oln_convol_fast_gaussian_derivative.png ** - ** \todo FIXME: Make the example use the good function. + ** \warning If sigma is big enough, the result may differ a + ** little bit with the -O3 flag of g++. */ template <class I, class BE> inline oln_concrete_type(I) @@ -236,7 +248,7 @@ ** ** im_type im1(oln::load(IMG_IN "lena256.pgm")); ** - ** oln::save(oln::convol::fast::gaussian(im1, 2, + ** oln::save(oln::convol::fast::gaussian_second_derivative(im1, 2, ** oln::value_behavior<oln_value_type_(im_type)>(0)), ** IMG_OUT "oln_convol_fast_gaussian_second_derivative.pbm"); ** return 0; @@ -249,7 +261,8 @@ ** \image html oln_convol_fast_gaussian_second_derivative.png ** \image latex oln_convol_fast_gaussian_second_derivative.png ** - ** \todo FIXME: Make the example use the good function. + ** \warning If sigma is big enough, the result may differ a + ** little bit with the -O3 flag of g++. */ template <class I, class BE> inline oln_concrete_type(I) @@ -297,6 +310,9 @@ ** => ** \image html oln_convol_fast_gaussian_default.png ** \image latex oln_convol_fast_gaussian_default.png + ** + ** \warning If sigma is big enough, the result may differ a + ** little bit with the -O3 flag of g++. */ template <class I> inline oln_concrete_type(I) @@ -332,7 +348,7 @@ ** ** im_type im1(oln::load(IMG_IN "lena256.pgm")); ** - ** oln::save(oln::convol::fast::gaussian(im1, 2), + ** oln::save(oln::convol::fast::gaussian_derivative(im1, 2), ** IMG_OUT "oln_convol_fast_gaussian_derivative_default.pbm"); ** return 0; ** } @@ -344,7 +360,8 @@ ** \image html oln_convol_fast_gaussian_derivative_default.png ** \image latex oln_convol_fast_gaussian_derivative_default.png ** - ** \todo FIXME: Make the example use the good function. + ** \warning If sigma is big enough, the result may differ a + ** little bit with the -O3 flag of g++. */ template <class I> inline oln_concrete_type(I) @@ -380,7 +397,7 @@ ** ** im_type im1(oln::load(IMG_IN "lena256.pgm")); ** - ** oln::save(oln::convol::fast::gaussian(im1, 2), + ** oln::save(oln::convol::fast::gaussian_second_derivative(im1, 2), ** IMG_OUT "oln_convol_fast_gaussian_second_derivative_default.pbm"); ** return 0; ** } @@ -392,7 +409,8 @@ ** \image html oln_convol_fast_gaussian_second_derivative_default.png ** \image latex oln_convol_fast_gaussian_second_derivative_default.png ** - ** \todo FIXME: Make the example use the good function. + ** \warning If sigma is big enough, the result may differ a + ** little bit with the -O3 flag of g++. */ template <class I> inline oln_concrete_type(I) Index: olena/oln/convol/fast_gaussian.hxx --- olena/oln/convol/fast_gaussian.hxx Mon, 15 Mar 2004 17:40:54 +0100 van-vl_n (oln/25_fast_gauss 1.7.1.8.1.9 600) +++ olena/oln/convol/fast_gaussian.hxx Mon, 29 Mar 2004 15:26:31 +0200 palma_g (oln/25_fast_gauss 1.7.1.8.1.9 640) @@ -104,7 +104,6 @@ - c.d[1]*tmp1[2] - c.d[2]*tmp1[1] - c.d[3]*tmp1[0]; - oln_point_type(I) current(start + d + d + d + d); for (coord i = 4; i < len; ++i) { @@ -170,7 +169,16 @@ struct gaussian_ {}; - // Gaussian filter for 1D images + /*! Gaussian filter for 1D images + ** + ** gaussian_ specialization for 1D images. + ** + ** \param I Exact type of the image. + ** \param F Type of coefficients. + ** + ** \arg img Image to process. + ** \arg coef Coefficients to use. + */ template<> struct gaussian_<1> { @@ -189,7 +197,16 @@ }; - // Gaussian filter for 2D images + /*! Gaussian filter for 2D images + ** + ** gaussian_ specialization for 2D images. + ** + ** \param I Exact type of the image. + ** \param F Type of coefficients. + ** + ** \arg img Image to process. + ** \arg coef Coefficients to use. + */ template<> struct gaussian_<2> { @@ -215,7 +232,16 @@ } }; - // Gaussian filter for 3D images + /*! Gaussian filter for 3D images + ** + ** gaussian_ specialization for 3D images. + ** + ** \param I Exact type of the image. + ** \param F Type of coefficients. + ** + ** \arg img Image to process. + ** \arg coef Coefficients to use. + */ template<> struct gaussian_<3> { Index: olena/tests/convol/tests/md5_convol_fast_gaussian_hh5 --- olena/tests/convol/tests/md5_convol_fast_gaussian_hh5 Tue, 23 Mar 2004 14:27:20 +0100 palma_g (oln/k/23_md5_convol 1.1 600) +++ olena/tests/convol/tests/md5_convol_fast_gaussian_hh5 Mon, 29 Mar 2004 15:33:06 +0200 palma_g (oln/k/23_md5_convol 1.1 644) @@ -10,17 +10,17 @@ int main() { - oln::utils::key::value_type data_key[16] = {0x70, 0x53, 0xa8, 0x84, 0x81, - 0x4, 0xbc, 0x30, 0x71, 0x23, - 0x78, 0xba, 0xed, 0x6, 0xc6, - 0x9e}; + oln::utils::key::value_type data_key[16] = {0xfc, 0xd6, 0xbc, 0xb5, 0x6c, + 0x16, 0x89, 0xfc, 0xef, 0x28, + 0xb5, 0x7c, 0x22, 0x47, 0x5b, + 0xad}; oln::utils::key key(data_key); typedef oln::image2d<ntg::int_u8> im_type; im_type im1(oln::load(rdata("lena256.pgm"))); - if (oln::utils::md5(oln::convol::fast::gaussian(im1, 2)) == key) + if (oln::utils::md5(oln::convol::fast::gaussian_second_derivative(im1, 0.5)) == key) std::cout << "OK" << std::endl; else { Index: olena/tests/convol/tests/md5_convol_fast_gaussian_hh4 --- olena/tests/convol/tests/md5_convol_fast_gaussian_hh4 Tue, 23 Mar 2004 14:27:20 +0100 palma_g (oln/k/24_md5_convol 1.1 600) +++ olena/tests/convol/tests/md5_convol_fast_gaussian_hh4 Mon, 29 Mar 2004 15:32:39 +0200 palma_g (oln/k/24_md5_convol 1.1 644) @@ -10,14 +10,17 @@ int main() { -oln::utils::key::value_type data_key[16] = {0x56, 0x10, 0x9f, 0xa9, 0xec, 0x89, 0x99, 0x14, 0x42, 0x4a, 0xcf, 0x42, 0xa9, 0xa2, 0xe3, 0xf4}; + oln::utils::key::value_type data_key[16] = {0x4d, 0xbc, 0xd, 0x2, 0x73, + 0x93, 0xf4, 0x6, 0xbc, 0x5c, + 0x55, 0xac, 0x6, 0xb, 0x67, + 0x9}; oln::utils::key key(data_key); - typedef oln::image2d<ntg:: -int_u8> im_type; + typedef oln::image2d<ntg::int_u8> im_type; im_type im1(oln::load(rdata("lena256.pgm"))); - if (oln::utils::md5(oln::convol::fast::gaussian(im1, 2)) == key) + + if (oln::utils::md5(oln::convol::fast::gaussian_derivative(im1, 0.5)) == key) std::cout << "OK" << std::endl; else { Index: olena/tests/convol/tests/md5_convol_fast_gaussian_hh3 --- olena/tests/convol/tests/md5_convol_fast_gaussian_hh3 Tue, 23 Mar 2004 14:27:20 +0100 palma_g (oln/k/25_md5_convol 1.1 600) +++ olena/tests/convol/tests/md5_convol_fast_gaussian_hh3 Mon, 29 Mar 2004 15:32:05 +0200 palma_g (oln/k/25_md5_convol 1.1 644) @@ -10,14 +10,16 @@ int main() { -oln::utils::key::value_type data_key[16] = {0x56, 0x10, 0x9f, 0xa9, 0xec, 0x89, 0x99, 0x14, 0x42, 0x4a, 0xcf, 0x42, 0xa9, 0xa2, 0xe3, 0xf4}; - + oln::utils::key::value_type data_key[16] = {0xf5, 0x37, 0xe0, 0xad, 0x5f, + 0xcc, 0x7f, 0x44, 0x32, 0x53, + 0x2a, 0x37, 0x80, 0x38, 0x3f, + 0xbb}; oln::utils::key key(data_key); - typedef oln::image2d<ntg:: -int_u8> im_type; + typedef oln::image2d<ntg::int_u8> im_type; im_type im1(oln::load(rdata("lena256.pgm"))); - if (oln::utils::md5(oln::convol::fast::gaussian(im1, 2)) == key) + + if (oln::utils::md5(oln::convol::fast::gaussian(im1, 0.5)) == key) std::cout << "OK" << std::endl; else { Index: olena/tests/convol/tests/md5_convol_fast_gaussian_hh2 --- olena/tests/convol/tests/md5_convol_fast_gaussian_hh2 Tue, 23 Mar 2004 14:27:20 +0100 palma_g (oln/k/26_md5_convol 1.1 600) +++ olena/tests/convol/tests/md5_convol_fast_gaussian_hh2 Mon, 29 Mar 2004 15:31:37 +0200 palma_g (oln/k/26_md5_convol 1.1 644) @@ -10,14 +10,20 @@ int main() { -oln::utils::key::value_type data_key[16] = {0xc, 0xec, 0xc0, 0x4c, 0xf4, 0xd1, 0xab, 0xef, 0x1c, 0xd3, 0x9c, 0xa7, 0x4d, 0xd7, 0x28, 0x92}; + oln::utils::key::value_type data_key[16] = {0xfc, 0xd6, 0xbc, 0xb5, 0x6c, + 0x16, 0x89, 0xfc, 0xef, 0x28, + 0xb5, 0x7c, 0x22, 0x47, 0x5b, + 0xad}; oln::utils::key key(data_key); - typedef oln::image2d<ntg:: -int_u8> im_type; + typedef oln::image2d<ntg::int_u8> im_type; im_type im1(oln::load(rdata("lena256.pgm"))); - if (oln::utils::md5(oln::convol::fast::gaussian(im1, 2, oln::value_behavior<oln_value_type_(im_type)>(0))) == key) + + if (oln::utils::md5(oln::convol::fast::gaussian_second_derivative(im1, + 0.5, + oln::value_behavior<oln_value_type_(im_type)>(0))) + == key) std::cout << "OK" << std::endl; else { Index: olena/tests/convol/tests/md5_convol_fast_gaussian_hh1 --- olena/tests/convol/tests/md5_convol_fast_gaussian_hh1 Tue, 23 Mar 2004 14:27:20 +0100 palma_g (oln/k/27_md5_convol 1.1 600) +++ olena/tests/convol/tests/md5_convol_fast_gaussian_hh1 Mon, 29 Mar 2004 15:31:03 +0200 palma_g (oln/k/27_md5_convol 1.1 644) @@ -10,19 +10,26 @@ int main() { -oln::utils::key::value_type data_key[16] = {0xc, 0xec, 0xc0, 0x4c, 0xf4, 0xd1, 0xab, 0xef, 0x1c, 0xd3, 0x9c, 0xa7, 0x4d, 0xd7, 0x28, 0x92}; + oln::utils::key::value_type data_key[16] = {0x0, 0x8f, 0x63, 0xff, 0xca, + 0xae, 0xb5, 0xdb, 0x32, 0xf5, + 0x38, 0xfd, 0x13, 0x11, 0x67, + 0x9c}; oln::utils::key key(data_key); - typedef oln::image2d<ntg:: -int_u8> im_type; + typedef oln::image2d<ntg::int_u8> im_type; im_type im1(oln::load(rdata("lena256.pgm"))); - if (oln::utils::md5(oln::convol::fast::gaussian(im1, 2, oln::value_behavior<oln_value_type_(im_type)>(0))) == key) + + if (oln::utils::md5(oln::convol::fast::gaussian_derivative(im1, + 0.5, + oln::value_behavior<oln_value_type_(im_type)>(0))) + == key) std::cout << "OK" << std::endl; else { std::cout << "FAIL" << std::endl; return true; } + return 0; } Index: olena/tests/convol/tests/md5_convol_fast_gaussian_hh0 --- olena/tests/convol/tests/md5_convol_fast_gaussian_hh0 Tue, 23 Mar 2004 14:27:20 +0100 palma_g (oln/k/28_md5_convol 1.1 600) +++ olena/tests/convol/tests/md5_convol_fast_gaussian_hh0 Mon, 29 Mar 2004 15:33:16 +0200 palma_g (oln/k/28_md5_convol 1.1 644) @@ -10,14 +10,20 @@ int main() { -oln::utils::key::value_type data_key[16] = {0xc, 0xec, 0xc0, 0x4c, 0xf4, 0xd1, 0xab, 0xef, 0x1c, 0xd3, 0x9c, 0xa7, 0x4d, 0xd7, 0x28, 0x92}; + oln::utils::key::value_type data_key[16] = {0xcd, 0x87, 0xa8, 0x95, 0xa3, + 0x44, 0x2f, 0x0, 0xc4, 0x39, + 0x30, 0x77, 0x67, 0xa8, 0xb7, + 0x36}; oln::utils::key key(data_key); - typedef oln::image2d<ntg:: -int_u8> im_type; + typedef oln::image2d<ntg::int_u8> im_type; im_type im1(oln::load(rdata("lena256.pgm"))); - if (oln::utils::md5(oln::convol::fast::gaussian(im1, 2, oln::value_behavior<oln_value_type_(im_type)>(0))) == key) + + if (oln::utils::md5(oln::convol::fast::gaussian(im1, + 0.5, + oln::value_behavior<oln_value_type_(im_type)>(0))) + == key) std::cout << "OK" << std::endl; else { -- Giovanni Palma EPITA - promo 2005 - membre d'EpX - LRDE Mob. : +33 (0)6 60 97 31 74