基于VTK库，用python语言写一个ICP的demo，

以下是一个基于VTK库，用Python语言实现ICP的简单示例代码，假设已经有两个点云`sourceCloud.vtk`和`targetCloud.vtk`，需要将`sourceCloud`对齐到`targetCloud`：

import vtk

# 读取源点云和目标点云
sourceReader = vtk.vtkPolyDataReader()
sourceReader.SetFileName("sourceCloud.vtk")
sourceReader.Update()

targetReader = vtk.vtkPolyDataReader()
targetReader.SetFileName("targetCloud.vtk")
targetReader.Update()

# ICP变换
icp = vtk.vtkIterativeClosestPointTransform()
icp.SetSource(sourceReader.GetOutput())
icp.SetTarget(targetReader.GetOutput())
icp.GetLandmarkTransform().SetModeToRigidBody()
icp.SetMaximumNumberOfIterations(100)
icp.StartByMatchingCentroidsOn()
icp.Update()

# 应用变换到源点云
transformed = vtk.vtkPolyData()
transformed.ShallowCopy(sourceReader.GetOutput())
transform = vtk.vtkTransform()
transform.SetMatrix(icp.GetMatrix())
transformFilter = vtk.vtkTransformPolyDataFilter()
transformFilter.SetTransform(transform)
transformFilter.SetInputData(transformed)
transformFilter.Update()

# 输出变换后的点云
writer = vtk.vtkPolyDataWriter()
writer.SetFileName("transformedCloud.vtk")
writer.SetInputData(transformFilter.GetOutput())
writer.Write()

该代码与C++版本类似，使用`vtkIterativeClosestPointTransform`实现ICP变换，并通过`vtkTransform`将变换应用到源点云上。最后将变换后的点云输出到文件中。