Skip to content

AlignTwoPolyDatas

vtk-examples/Python/PolyData/AlignTwoPolyDatas

Description

This example shows how to align two vtkPolyData's. Typically, the two datasets are related. For example, aligning a CT head isosurface with an MRI head isosurface of the same patient. Or two steps in a time series of an evolving surface. These cases usually reside in the same coordinate system, and the initial alignment is "close" to the desired results.

Another case is when the two datasets are from the "same" family of objects — for example, running the example with two types of sharks that exist in different coordinate systems.

The algorithm proceeds as follows:

  1. Read the two vtkPolyData's that exist in the example's command line. The first file contains the source vtkPolyData to be aligned with the second file's vtkPolyData called the target. Another naming convention is moving and fixed.
  2. Compute a measure of fit of the two original files. We use the recently added vtkHausdorffDistancePointSetFilter to compute the measure. See Hausdorff Distance.
  3. Align the bounding boxes of the two datasets. Here we use a vtkOBBTree locator to create oriented bounding boxes. See Oriented Bounding Boxes. Use the bounding box corner coordinates to create source and target vtkLandmarkTransform's. vtkTransformPolyData uses this transform to create a new source vtkPolyData. Since the orientations of the bounding boxes may differ, the AlignBoundingBoxes function tries ten different rotations. For each rotation, it computes the Hausdorff distance between the target's OBB corners and the transformed source's OBB corners. Finally, transform the original source using the smallest distance.
  4. Improve the alignment with vtkIterativeClosestPointTransform with a RigidBody transform. Compute the distance metric again.
  5. Using the transform that has the best distance metric, do a final, and display the source and target vtkPolyData's.

Info

The example is run with src/Testing/Data/thingiverse/Grey_Nurse_Shark.stl and src/Testing/Data/shark.ply

Info

You may need to orient the target using vtkTransformPolyDataFilter to get a better fit, for example when using src/Testing/Data/thingiverse/Grey_Nurse_Shark.stl and src/Testing/Data/greatWhite.stl, uncommenting the two rotations in the vtkTransform will provide an excellent alignment.

Other languages

See (Cxx)

Question

If you have a question about this example, please use the VTK Discourse Forum

Code

AlignTwoPolyDatas.py

#!/usr/bin/env python
# -*- coding: utf-8 -*-

import vtk


def get_program_parameters():
    import argparse
    description = 'How to align two vtkPolyData\'s.'
    epilogue = '''

    '''
    parser = argparse.ArgumentParser(description=description, epilog=epilogue,
                                     formatter_class=argparse.RawTextHelpFormatter)
    parser.add_argument('src_fn', help='The polydata source file name,e.g. Grey_Nurse_Shark.stl.')
    parser.add_argument('tgt_fn', help='The polydata target file name, e.g. shark.ply.')

    args = parser.parse_args()

    return args.src_fn, args.tgt_fn


def main():
    colors = vtk.vtkNamedColors()

    src_fn, tgt_fn = get_program_parameters()
    print('Loading source:', src_fn)
    sourcePolyData = ReadPolyData(src_fn)
    # Save the source polydata in case the align does not improve
    # segmentation
    originalSourcePolyData = vtk.vtkPolyData()
    originalSourcePolyData.DeepCopy(sourcePolyData)

    print('Loading target:', tgt_fn)
    targetPolyData = ReadPolyData(tgt_fn)

    # If the target orientation is markedly different,
    # you may need to apply a transform to orient the
    # target with the source.
    # For example, when using Grey_Nurse_Shark.stl as the source and
    # greatWhite.stl as the target, you need to uncomment the following
    # two rotations.
    trnf = vtk.vtkTransform()
    # trnf.RotateX(90)
    # trnf.RotateY(-90)
    tpd = vtk.vtkTransformPolyDataFilter()
    tpd.SetTransform(trnf)
    tpd.SetInputData(targetPolyData)
    tpd.Update()

    renderer = vtk.vtkRenderer()
    renderWindow = vtk.vtkRenderWindow()
    renderWindow.AddRenderer(renderer)
    interactor = vtk.vtkRenderWindowInteractor()
    interactor.SetRenderWindow(renderWindow)

    distance = vtk.vtkHausdorffDistancePointSetFilter()
    distance.SetInputData(0, tpd.GetOutput())
    distance.SetInputData(1, sourcePolyData)
    distance.Update()

    distanceBeforeAlign = distance.GetOutput(0).GetFieldData().GetArray('HausdorffDistance').GetComponent(0, 0)

    # Get initial alignment using oriented bounding boxes
    AlignBoundingBoxes(sourcePolyData, tpd.GetOutput())

    distance.SetInputData(0, tpd.GetOutput())
    distance.SetInputData(1, sourcePolyData)
    distance.Modified()
    distance.Update()
    distanceAfterAlign = distance.GetOutput(0).GetFieldData().GetArray('HausdorffDistance').GetComponent(0, 0)

    bestDistance = min(distanceBeforeAlign, distanceAfterAlign)

    if distanceAfterAlign > distanceBeforeAlign:
        sourcePolyData.DeepCopy(originalSourcePolyData)

    # Refine the alignment using IterativeClosestPoint
    icp = vtk.vtkIterativeClosestPointTransform()
    icp.SetSource(sourcePolyData)
    icp.SetTarget(tpd.GetOutput())
    icp.GetLandmarkTransform().SetModeToRigidBody()
    icp.SetMaximumNumberOfLandmarks(100)
    icp.SetMaximumMeanDistance(.00001)
    icp.SetMaximumNumberOfIterations(500)
    icp.CheckMeanDistanceOn()
    icp.StartByMatchingCentroidsOn()
    icp.Update()

    #  print(icp)

    lmTransform = icp.GetLandmarkTransform()
    transform = vtk.vtkTransformPolyDataFilter()
    transform.SetInputData(sourcePolyData)
    transform.SetTransform(lmTransform)
    transform.SetTransform(icp)
    transform.Update()

    distance.SetInputData(0, tpd.GetOutput())
    distance.SetInputData(1, transform.GetOutput())
    distance.Update()

    distanceAfterICP = distance.GetOutput(0).GetFieldData().GetArray('HausdorffDistance').GetComponent(0, 0)

    if distanceAfterICP < bestDistance:
        bestDistance = distanceAfterICP

    print(
        'Distance before, after align, after ICP, min: {:0.5f}, {:0.5f}, {:0.5f}, {:0.5f}'.format(distanceBeforeAlign,
                                                                                                  distanceAfterAlign,
                                                                                                  distanceAfterICP,
                                                                                                  bestDistance))
    # Select
    sourceMapper = vtk.vtkDataSetMapper()
    if bestDistance == distanceBeforeAlign:
        sourceMapper.SetInputData(originalSourcePolyData)
        print('Using original alignment')
    elif bestDistance == distanceAfterAlign:
        sourceMapper.SetInputData(sourcePolyData)
        print('Using alignment by OBB')
    else:
        sourceMapper.SetInputConnection(transform.GetOutputPort())
        print('Using alignment by ICP')
    sourceMapper.ScalarVisibilityOff()

    sourceActor = vtk.vtkActor()
    sourceActor.SetMapper(sourceMapper)
    sourceActor.GetProperty().SetOpacity(0.6)
    sourceActor.GetProperty().SetDiffuseColor(
        colors.GetColor3d('White'))
    renderer.AddActor(sourceActor)

    targetMapper = vtk.vtkDataSetMapper()
    targetMapper.SetInputData(tpd.GetOutput())
    targetMapper.ScalarVisibilityOff()

    targetActor = vtk.vtkActor()
    targetActor.SetMapper(targetMapper)
    targetActor.GetProperty().SetDiffuseColor(
        colors.GetColor3d('Tomato'))
    renderer.AddActor(targetActor)

    renderWindow.AddRenderer(renderer)
    renderer.SetBackground(colors.GetColor3d("sea_green_light"))
    renderer.UseHiddenLineRemovalOn()

    renderWindow.SetSize(640, 480)
    renderWindow.Render()
    renderWindow.SetWindowName('AlignTwoPolyDatas')
    renderWindow.Render()
    interactor.Start()


def ReadPolyData(file_name):
    import os
    path, extension = os.path.splitext(file_name)
    extension = extension.lower()
    if extension == ".ply":
        reader = vtk.vtkPLYReader()
        reader.SetFileName(file_name)
        reader.Update()
        poly_data = reader.GetOutput()
    elif extension == ".vtp":
        reader = vtk.vtkXMLpoly_dataReader()
        reader.SetFileName(file_name)
        reader.Update()
        poly_data = reader.GetOutput()
    elif extension == ".obj":
        reader = vtk.vtkOBJReader()
        reader.SetFileName(file_name)
        reader.Update()
        poly_data = reader.GetOutput()
    elif extension == ".stl":
        reader = vtk.vtkSTLReader()
        reader.SetFileName(file_name)
        reader.Update()
        poly_data = reader.GetOutput()
    elif extension == ".vtk":
        reader = vtk.vtkpoly_dataReader()
        reader.SetFileName(file_name)
        reader.Update()
        poly_data = reader.GetOutput()
    elif extension == ".g":
        reader = vtk.vtkBYUReader()
        reader.SetGeometryFileName(file_name)
        reader.Update()
        poly_data = reader.GetOutput()
    else:
        # Return a None if the extension is unknown.
        poly_data = None
    return poly_data


def AlignBoundingBoxes(source, target):
    # Use OBBTree to create an oriented bounding box for target and source
    sourceOBBTree = vtk.vtkOBBTree()
    sourceOBBTree.SetDataSet(source)
    sourceOBBTree.SetMaxLevel(1)
    sourceOBBTree.BuildLocator()

    targetOBBTree = vtk.vtkOBBTree()
    targetOBBTree.SetDataSet(target)
    targetOBBTree.SetMaxLevel(1)
    targetOBBTree.BuildLocator()

    sourceLandmarks = vtk.vtkPolyData()
    sourceOBBTree.GenerateRepresentation(0, sourceLandmarks)

    targetLandmarks = vtk.vtkPolyData()
    targetOBBTree.GenerateRepresentation(0, targetLandmarks)

    lmTransform = vtk.vtkLandmarkTransform()
    lmTransform.SetModeToSimilarity()
    lmTransform.SetTargetLandmarks(targetLandmarks.GetPoints())
    # lmTransformPD = vtk.vtkTransformPolyDataFilter()
    bestDistance = vtk.VTK_DOUBLE_MAX
    bestPoints = vtk.vtkPoints()
    bestDistance = BestBoundingBox(
        "X",
        target,
        source,
        targetLandmarks,
        sourceLandmarks,
        bestDistance,
        bestPoints)
    bestDistance = BestBoundingBox(
        "Y",
        target,
        source,
        targetLandmarks,
        sourceLandmarks,
        bestDistance,
        bestPoints)
    bestDistance = BestBoundingBox(
        "Z",
        target,
        source,
        targetLandmarks,
        sourceLandmarks,
        bestDistance,
        bestPoints)

    lmTransform.SetSourceLandmarks(bestPoints)
    lmTransform.Modified()

    transformPD = vtk.vtkTransformPolyDataFilter()
    transformPD.SetInputData(source)
    transformPD.SetTransform(lmTransform)
    transformPD.Update()

    source.DeepCopy(transformPD.GetOutput())

    return


def BestBoundingBox(axis, target, source, targetLandmarks, sourceLandmarks, bestDistance, bestPoints):
    distance = vtk.vtkHausdorffDistancePointSetFilter()
    testTransform = vtk.vtkTransform()
    testTransformPD = vtk.vtkTransformPolyDataFilter()
    lmTransform = vtk.vtkLandmarkTransform()
    lmTransformPD = vtk.vtkTransformPolyDataFilter()

    lmTransform.SetModeToSimilarity()
    lmTransform.SetTargetLandmarks(targetLandmarks.GetPoints())

    sourceCenter = sourceLandmarks.GetCenter()

    delta = 90.0
    for i in range(0, 4):
        angle = delta * i
        # Rotate about center
        testTransform.Identity()
        testTransform.Translate(sourceCenter[0], sourceCenter[1], sourceCenter[2])
        if axis == "X":
            testTransform.RotateX(angle)
        elif axis == "Y":
            testTransform.RotateY(angle)
        else:
            testTransform.RotateZ(angle)
        testTransform.Translate(-sourceCenter[0], -sourceCenter[1], -sourceCenter[2])

        testTransformPD.SetTransform(testTransform)
        testTransformPD.SetInputData(sourceLandmarks)
        testTransformPD.Update()

        lmTransform.SetSourceLandmarks(testTransformPD.GetOutput().GetPoints())
        lmTransform.Modified()

        lmTransformPD.SetInputData(source)
        lmTransformPD.SetTransform(lmTransform)
        lmTransformPD.Update()

        distance.SetInputData(0, target)
        distance.SetInputData(1, lmTransformPD.GetOutput())
        distance.Update()

        testDistance = distance.GetOutput(0).GetFieldData().GetArray("HausdorffDistance").GetComponent(0, 0)
        if testDistance < bestDistance:
            bestDistance = testDistance
            bestPoints.DeepCopy(testTransformPD.GetOutput().GetPoints())

    return bestDistance


if __name__ == '__main__':
    main()