tf_inpainting.cc

Go to the documentation of this file.

#include <algorithm>
#include <exception>
#include <functional>
#include <iostream>
#include <random>
#include <vector>
#include <ctime>
#include <catch2/catch_all.hpp>
 
#include "sopt/imaging_forward_backward.h"
#include "sopt/tf_non_diff_function.h"
#include "sopt/logging.h"
#include "sopt/maths.h"
#include "sopt/relative_variation.h"
#include "sopt/sampling.h"
#include "sopt/types.h"
#include "sopt/utilities.h"
 
// This header is not part of the installed sopt interface
// It is only present in tests
#include "tools_for_tests/directories.h"
#include "tools_for_tests/tiffwrappers.h"
 
// \min_{x} ||\Psi^Tx||_1 \quad \mbox{s.t.} \quad ||y - Ax||_2 < \epsilon and x \geq 0
 
using Scalar = double;
using Vector = sopt::Vector<Scalar>;
using Matrix = sopt::Matrix<Scalar>;
using Image = sopt::Image<Scalar>;
 
TEST_CASE("Inpainting"){
  extern std::unique_ptr<std::mt19937_64> mersenne;
  const std::string input = "cameraman256";
 
  const std::string model_path = std::string(sopt::tools::models_directory() + "/snr_15_model_dynamic.onnx");
 
  const Image image = sopt::tools::read_standard_tiff(input);
 
  sopt::t_uint nmeasure = std::floor(0.5 * image.size());
  sopt::LinearTransform<Vector> const sampling =
      sopt::linear_transform<Scalar>(sopt::Sampling(image.size(), nmeasure, *mersenne));
 
  const Vector y0 = sampling * Vector::Map(image.data(), image.size());
  constexpr auto snr = 30.0;
  const auto sigma = y0.stableNorm() / std::sqrt(y0.size()) * std::pow(10.0, -(snr / 20.0));
  const auto epsilon = std::sqrt(nmeasure + 2 * std::sqrt(y0.size())) * sigma;
 
  std::normal_distribution<> gaussian_dist(0, sigma);
  Vector y(y0.size());
  for (sopt::t_int i = 0; i < y0.size(); i++) y(i) = y0(i) + gaussian_dist(*mersenne);
 
  Eigen::VectorXd dirty_image = sampling.adjoint() * y;
 
  sopt::t_real constexpr regulariser_strength = 100;
  sopt::t_real const beta = sigma * sigma * 0.5;
 
  auto fb = sopt::algorithm::ImagingForwardBackward<Scalar>(y);
  fb.itermax(500)
    .step_size(beta)    // stepsize
    .sigma(sigma)  // sigma
    .regulariser_strength(regulariser_strength)  // regularisation paramater
    .fista(false)  // switch to use FISTA algorithm in Forward Backward algorithm, should be false if using learned TF model
    .relative_variation(1e-3)
    .residual_tolerance(0)
    .tight_frame(true)
    .Phi(sampling);
 
  // Create a shared pointer to an instance of the TFGProximal class
  auto gp = std::make_shared<sopt::algorithm::TFGProximal<Scalar>>(model_path);
 
  // Inject it into the ImagingForwardBackward object
  fb.g_function(gp);
 
  auto const diagnostic = fb();
 
  CHECK(diagnostic.good);
  CHECK(diagnostic.niters < 500);
 
  // compare input image to cleaned output image
  // calculate mean squared error sum_i ( ( x_true(i) - x_est(i) ) **2 )
  // check this is less than the number of pixels * 0.01
  sopt::utilities::write_tiff(Matrix::Map(diagnostic.x.data(), image.rows(), image.cols()),
                                "tf_reconstruction.tiff");
  sopt::utilities::write_tiff(Matrix::Map(dirty_image.data(), image.rows(), image.cols()),
                                "tf_dirty.tiff");
  Eigen::Map<const Eigen::VectorXd> flat_image(image.data(), image.size());
  auto mse = (flat_image - diagnostic.x).array().square().sum() / image.size();
  CAPTURE(mse);
  SOPT_HIGH_LOG("MSE: {}", mse);
  CHECK(mse < 0.01);
}