[Olena] oln 10.56: Gaussian correction.

Index: integre/ChangeLog from Giovanni Palma &lt;giovanni(a)lrde.epita.fr&gt; * ntg/utils/cast.hh: Make force use a single cast instead of delegate its work to an unsafe type. Index: olena/ChangeLog from Giovanni Palma &lt;giovanni(a)lrde.epita.fr&gt; * oln/convol/fast_gaussian.hxx: Make the gaussian take care of the borders. Index: olena/oln/convol/fast_gaussian.hxx --- olena/oln/convol/fast_gaussian.hxx Thu, 07 Aug 2003 02:08:21 +0200 david (oln/25_fast_gauss 1.7.1.8.1.4 640) +++ olena/oln/convol/fast_gaussian.hxx Wed, 28 Jan 2004 15:34:38 +0100 palma_g (oln/25_fast_gauss 1.7.1.8.1.4 640) @@ -159,9 +159,9 @@ // Apply on columns. recursivefilter_<float>(img, coef, - oln_point_type(I)(0), - oln_point_type(I)(img.ncols() - 1), - img.ncols(), + oln_point_type(I)(-img.border()), + oln_point_type(I)(img.ncols() - 1 + img.border()), + img.ncols() + 2 * img.border(), oln_dpoint_type(I)(1)); } }; @@ -175,21 +175,20 @@ void doit(abstract::image_with_dim<2, I>& img, const F& coef) { - // Apply on rows. for (coord j = 0; j < img.ncols(); ++j) recursivefilter_<float>(img, coef, - oln_point_type(I)(0, j), - oln_point_type(I)(img.nrows() - 1, j), - img.nrows(), + oln_point_type(I)(-img.border(), j), + oln_point_type(I)(img.nrows() - 1 + img.border(), j), + img.nrows() + 2 * img.border(), oln_dpoint_type(I)(1, 0)); // Apply on columns. for (coord i = 0; i < img.nrows(); ++i) recursivefilter_<float>(img, coef, - oln_point_type(I)(i, 0), - oln_point_type(I)(i, img.ncols() - 1), - img.ncols(), + oln_point_type(I)(i, -img.border()), + oln_point_type(I)(i, img.ncols() - 1 + img.border()), + img.ncols() + 2 * img.border(), oln_dpoint_type(I)(0, 1)); } }; @@ -206,27 +205,27 @@ for (coord j = 0; j < img.nrows(); ++j) for (coord k = 0; k < img.ncols(); ++k) recursivefilter_<float>(img, coef, - oln_point_type(I)(0, j, k), - oln_point_type(I)(img.nslices() - 1, j, k), - img.ncols(), + oln_point_type(I)(-img.border(), j, k), + oln_point_type(I)(img.nslices() - 1 + img.border(), j, k), + img.ncols() + 2 * img.border(), oln_dpoint_type(I)(1, 0, 0)); // Apply on rows. for (coord i = 0; i < img.nslices(); ++i) for (coord k = 0; k < img.ncols(); ++k) recursivefilter_<float>(img, coef, - oln_point_type(I)(i, 0, k), - oln_point_type(I)(i, img.nrows() - 1, k), - img.nrows(), + oln_point_type(I)(i, -img.border(), k), + oln_point_type(I)(i, img.nrows() - 1 + img.border(), k), + img.nrows() + 2 * img.border(), oln_dpoint_type(I)(0, 1, 0)); // Apply on columns. for (coord i = 0; i < img.nslices(); ++i) for (coord j = 0; j < img.nrows(); ++j) recursivefilter_<float>(img, coef, - oln_point_type(I)(i, j, 0), - oln_point_type(I)(i, j, img.ncols() - 1), - img.ncols(), + oln_point_type(I)(i, j, -img.border()), + oln_point_type(I)(i, j, img.ncols() - 1 + img.border()), + img.ncols() + 2 * img.border(), oln_dpoint_type(I)(0, 0, 1)); } }; @@ -235,7 +234,8 @@ typename mute<I, typename convoutput<C,B,oln_value_type(I)>::ret>::ret gaussian_common_(const convert::abstract::conversion<C,B>& c, const abstract::image<I>& in, - const F& coef) + const F& coef, + ntg::float_s sigma) { typename mute<I, ntg::float_s>::ret work_img(in.size()); @@ -243,6 +243,12 @@ for_all(it) work_img[it] = ntg::cast::force<ntg::float_s>(in[it]); + /* FIXME: relation between sigma and the border shouldn't + be linear, so when sigma is big enougth, the signal may + be parasitized by the non signal values. + */ + work_img.border_adapt_mirror(ntg::cast::round<coord>(5 * sigma)); + gaussian_<I::dim>::doit(work_img, coef); /* Convert the result image to the user-requested datatype. @@ -250,6 +256,7 @@ user expects a ntg::float_s image. */ typename mute<I, typename convoutput<C,B,oln_value_type(I)>::ret>::ret out_img(in.size()); + for_all(it) out_img[it] = c(work_img[it]); @@ -271,7 +278,7 @@ sigma, internal::recursivefilter_coef_<float>::DericheGaussian); - return internal::gaussian_common_(c, in, coef); + return internal::gaussian_common_(c, in, coef, sigma); } template <class C, class B, class I> @@ -288,7 +295,7 @@ internal::recursivefilter_coef_<float> ::DericheGaussianFirstDerivative); - return internal::gaussian_common_(c, in, coef); + return internal::gaussian_common_(c, in, coef, sigma); } template <class C, class B, class I> @@ -305,7 +312,7 @@ internal::recursivefilter_coef_<float> ::DericheGaussianSecondDerivative); - return internal::gaussian_common_(c, in, coef); + return internal::gaussian_common_(c, in, coef, sigma); } } // fast Index: integre/ntg/utils/cast.hh --- integre/ntg/utils/cast.hh Thu, 27 Nov 2003 11:26:27 +0100 burrus_n (oln/i/26_cast.hh 1.3.1.11 640) +++ integre/ntg/utils/cast.hh Wed, 28 Jan 2004 16:22:43 +0100 palma_g (oln/i/26_cast.hh 1.3.1.11 640) @@ -70,12 +70,14 @@ | force | `------*/ + // a cast is performed instead of a constructor (unsafe one) call + // because this last one may not be available with the good + // signature. template<class Tdest, class Tsrc> inline const Tdest force(const Tsrc& val) { - ntg_unsafe_type(Tdest) tmp (val); - return tmp; + return (Tdest)val; } /*------. @@ -115,13 +117,13 @@ { // FIXME: this code seems out of date. -#if 0 +#if 1 // KLUDGE: Cast the rounded value to Tdest::value_t before - // returning it as Tdest. Otherwise g++-3.0 complains there + // returning it as Tdest. Otherwise g++-3.* complains there // is no Tdest constructor taking a float argument. return (ntg_storage_type(Tdest)) round(val.exact()); #endif - return round(val.exact()); + // return round(val.exact()); } }; @@ -132,13 +134,13 @@ doit(const float_s& val) { // FIXME: this code seems out of date. -#if 0 +#if 1 // KLUDGE: Cast the rounded value to Tdest::value_t before - // returning it as Tdest. Otherwise g++-3.0 complains there + // returning it as Tdest. Otherwise g++-3.* complains there // is no Tdest constructor taking a float argument. - // return (ntg_storage_type(Tdest)) roundf(val); + return (ntg_storage_type(Tdest)) roundf(val); #endif - return roundf(val); + // return roundf(val); } }; -- Giovanni Palma EPITA - promo 2005 - membre d'EpX - LRDE Mob. : +33 (0)6 60 97 31 74