/*=========================================================================
 *
 *  Copyright NumFOCUS
 *
 *  Licensed under the Apache License, Version 2.0 (the "License");
 *  you may not use this file except in compliance with the License.
 *  You may obtain a copy of the License at
 *
 *         https://www.apache.org/licenses/LICENSE-2.0.txt
 *
 *  Unless required by applicable law or agreed to in writing, software
 *  distributed under the License is distributed on an "AS IS" BASIS,
 *  WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 *  See the License for the specific language governing permissions and
 *  limitations under the License.
 *
 *=========================================================================*/
#ifndef itkCannySegmentationLevelSetFunction_hxx
#define itkCannySegmentationLevelSetFunction_hxx

#include "itkGradientImageFilter.h"
#include "itkMultiplyImageFilter.h"
#include "itkImageRegionIterator.h"
#include "itkImageAlgorithm.h"

namespace itk
{
template <typename TImageType, typename TFeatureImageType>
void
CannySegmentationLevelSetFunction<TImageType, TFeatureImageType>::CalculateSpeedImage()
{
  // Create a distance transform to the canny edges
  this->CalculateDistanceImage();

  // Graft the distance transform into the Speed Image
  this->GetSpeedImage()->Graft(m_Distance->GetOutput());
}

template <typename TImageType, typename TFeatureImageType>
void
CannySegmentationLevelSetFunction<TImageType, TFeatureImageType>::CalculateAdvectionImage()
{
  typename GradientImageFilter<ImageType, ScalarValueType, ScalarValueType>::Pointer gradient =
    GradientImageFilter<ImageType, ScalarValueType, ScalarValueType>::New();

  using CovariantVectorImageType =
    typename GradientImageFilter<ImageType, ScalarValueType, ScalarValueType>::OutputImageType;

  typename MultiplyImageFilter<CovariantVectorImageType, ImageType, CovariantVectorImageType>::Pointer multiply =
    MultiplyImageFilter<CovariantVectorImageType, ImageType, CovariantVectorImageType>::New();

  // Create a distance transform to the canny edges
  this->CalculateDistanceImage();

  gradient->SetInput(m_Distance->GetOutput());
  gradient->Update();

  multiply->SetInput1(gradient->GetOutput());
  multiply->SetInput2(m_Distance->GetOutput());

  //  multiply->GraftOutput(dynamic_cast<CovariantVectorImageType
  // *>(this->GetAdvectionImage()));
  multiply->Update();

  // Copy output to Advection Image
  ImageAlgorithm::Copy(multiply->GetOutput(),
                       this->GetAdvectionImage(),
                       this->GetAdvectionImage()->GetRequestedRegion(),
                       this->GetAdvectionImage()->GetRequestedRegion());
}

template <typename TImageType, typename TFeatureImageType>
void
CannySegmentationLevelSetFunction<TImageType, TFeatureImageType>::CalculateDistanceImage()
{
  auto tempFeature = TFeatureImageType::New();

  // The minipipeline might muck with its feature image requested
  // region. The rest of the class relies on the feature image requested
  // region as specified by the original level set
  // filter. We make a temporary shallow copy of feature image to
  // build the distance image.
  tempFeature->Graft(this->GetFeatureImage());

  // AssignCannyInput either sets up a pipeline through the
  // CastImageFilter if TImageType != TFeatureImageType
  // or bypasses the Cast operation if TImageType == TFeatureType
  typename TImageType::Pointer junk;
  this->AssignCannyInput(tempFeature, junk);

  m_Canny->SetUpperThreshold(m_Threshold);
  m_Canny->SetVariance(m_Variance);
  m_Canny->SetMaximumError(0.01);

  m_Distance->SetInput(m_Canny->GetOutput());
  m_Distance->GetOutput()->SetRequestedRegion(this->GetSpeedImage()->GetRequestedRegion());
  m_Distance->Update();
}
} // end namespace itk

#endif