I am trying to use vtkImageReSlicer to extract a 2d slice from a 3d
vtkImageData object. But I can't seem to get the recipe right. Am I doing it right?
I am also a bit confused about ResliceAxes Matrix. Does it represent a cutting plane? If
I move the ReSliceAxes origin will it also move the cutting plane? When I
call Update on the vtkImageReSlicer, the program crashes. But when I don't
call it, the output is empty.
Here's what I have so far.
#my input is any vtkactor that contains a closed curve of type vtkPolyData
ShapePolyData = actor.GetMapper().GetInput()
boundingBox = ShapePolyData.GetBounds()
for i in range(0,6,2):
delta = boundingBox[i+1]-boundingBox[i]
newBoundingBox.append(boundingBox[i]-0.5*delta)
newBoundingBox.append(boundingBox[i+1]+0.5*delta)
voxelizer = vtk.vtkVoxelModeller()
voxelizer.SetInputData(ShapePolyData)
voxelizer.SetModelBounds(newBoundingBox)
voxelizer.SetScalarTypeToBit()
voxelizer.SetForegroundValue(1)
voxelizer.SetBackgroundValue(0)
voxelizer.Update()
VoxelModel =voxelizer.GetOutput()
ImageOrigin = VoxelModel.GetOrigin()
slicer = vtk.vtkImageReslice()
#Am I setting the cutting axis here. x axis set at 1,0,0 , y axis at 0,1,0 and z axis at 0,0,1
slicer.SetResliceAxesDirectionCosines(1,0,0,0,1,0,0,0,1)
#if I increase the z value, will the cutting plane move up?
slicer.SetResliceAxesOrigin(ImageOrigin[0],ImageOrigin[1],ImageOrigin[2])
slicer.SetInputData(VoxelModel)
slicer.SetInterpolationModeToLinear()
slicer.SetOutputDimensionality(2)
slicer.Update() #this makes the code crash
voxelSurface = vtk.vtkContourFilter()
voxelSurface.SetInputConnection(slicer.GetOutputPort())
voxelSurface.SetValue(0, .999)
voxelMapper = vtk.vtkPolyDataMapper()
voxelMapper.SetInputConnection(voxelSurface.GetOutputPort())
voxelActor = vtk.vtkActor()
voxelActor.SetMapper(voxelMapper)
Renderer.AddActor(voxelActor)
I have never used vtkImageReslice, but I have used vtkExtractVOI for vtkImageData, which allows you to achieve a similar result, I think. Here is your example modified with the latter, instead:
ImageOrigin = VoxelModel.GetOrigin()
slicer = vtk.vtkExtractVOI()
slicer.SetInputData(VoxelModel)
#With the setVOI method you can define which slice you want to extract
slicer.SetVOI(xmin, xmax, ymin, ymax, zslice, zslice)
slicer.SetSampleRate(1, 1, 1)
slicer.Update()
voxelSurface = vtk.vtkContourFilter()
voxelSurface.SetInputConnection(slicer.GetOutputPort())
voxelSurface.SetValue(0, .999)
voxelMapper = vtk.vtkPolyDataMapper()
voxelMapper.SetInputConnection(voxelSurface.GetOutputPort())
voxelActor = vtk.vtkActor()
voxelActor.SetMapper(voxelMapper)
Renderer.AddActor(voxelActor)
Related
I was making animation with manim and I tried to make line that one end is fixed, and one end moves along the arc. The way I wanted to solve this problem was to define each point with theta, and let the point move along the arc as it moves. However, even if the setta was changed, the line did not move, so the rotate was used, but the length of the line did not change, so the line deviated from the arc and failed. I wonder how to make the line move smoothly with the size of the angle when theta value is changed to an arbitrary value with theta as a parameter.
from manim import *
import numpy as np
class RotateVector(Scene):
def construct(self):
theta = 1/9*np.pi
semicircle = Arc(angle=PI, radius=4, stroke_color = GREY, arc_center=np.array([0, -1.5, 0]))
bottom_line = Line(start=np.array([-4.0,-1.5,0]), end=np.array([4.0,-1.5,0]), stroke_color = GREY)
dot_A = np.array([-4.0,-1.5,0])
dot_A_text = Text('A', font_size=35).next_to(dot_A, LEFT+DOWN, buff=0.05)
dot_P = np.array([4*np.cos(theta*2),4*np.sin(theta*2)-1.5,0])
dot_P_text = Text('P', font_size=35).next_to(dot_P, RIGHT+UP, buff=0.05)
line_AP = Line(start=np.array([-4.0,-1.5,0]), end=np.array([4*np.cos(2*theta),4*np.sin(2*theta)-1.5,0]), stroke_color = GREY)
self.play(Create(semicircle), Create(bottom_line), Create(dot_A_text))
self.play(Create(line_AP), Create(dot_P_text))
self.wait(3)
Your mobjects do not change because the dependency on Theta is not retained after initializing the mobject -- but it doesn't take too much to make it work!
The solution to your Problem is using a ValueTracker for the value of theta (to easily allow animating changing it) and updater functions for the mobjects you would like to change. In other words:
theta_vt = ValueTracker(theta)
dot_P_text.add_updater(lambda mobj: mobj.next_to([4*np.cos(theta_vt.get_value()*2),4*np.sin(theta_vt.get_value()*2)-1.5,0]))
line_AP.add_updater(lambda mobj: mobj.put_start_and_end_on([-4, 1.5, 0], [4*np.cos(theta_vt.get_value()*2),4*np.sin(theta_vt.get_value()*2)-1.5,0]))
self.play(theta_vt.animate.set_value(2*PI))
You could even avoid the value tracker and use an auxiliary (empty) mobject for positioning:
P = Mobject().move_to([4*np.cos(theta*2),4*np.sin(theta*2)-1.5,0])
and then change the updaters to use the position of P instead of theta, and make the point move along the desired arc. (That point actually is the one that becomes slightly more complicated, while the updaters simplify a bit.)
I have a set of 3d points (I generate the positions of planets and moons in a stellar system from Keplers equations) I have the coordinates of all points as x,y,z, where the central star is 0,0,0. The code to produce the points works perfectly.
However, at the moment I plot a visualisation of this system from above - so I just ignore the z component for the purposes of visualisation and plot the x and y to the canvas as-is. This works as intended.
How would I generate x and y coordinates for plotting to the canvas that take into account the z coordinate, so that I can plot a view from another angle apart from directly above?
The only library I can use apart from the standard one would be numpy. I cannot use Matplotlib.
edit thanks to the comments I can now clarify with some psudocode.
Assume I have a bunch of points that have an xyz position.
What I currently do:
canvas.plot(point.x)
canvas.plot(point.y)
ignoring point z - so that it is as if all z's are 0 and it is viewed from 'above'
So that I can use my current plotting code - which takes into account scale and offsets to do with the canvas, I need new x and y coordinates that are as if the view is from another angle other than 'above'.
It seems from the helpful comments what I have to do is rotate the whole coordinate system so that it has a new z axis that is a result of a rotation of the whole system about the x and y axis.
Something like the following psudocode would do.
def rotate_about_axis(x_rotation_degrees, y_rotation_degrees, point.x, point.y, point.z):
new_plot_x = canvas_point_to_plot after magic code to rotate coordinates x_rotation_degrees about x axis
new_plot_y = canvas_point_to_plot after magic code to rotate coordinates y_rotation_degrees about y axis
return new_plot_x, new_plot_y
Then I could apply this to all the points I plot.
How would I do this in python?
I have come up with an answer, I hope it helps someone.
import numpy, math
def rotate_about_axis(x_rotation_degrees, y_rotation_degrees, point_x, point_y, point_z):
xrads = math.radians(x_rotation_degrees)
yrads = math.radians(y_rotation_degrees)
rotation = [xrads, yrads, 0]
rotation_angle = numpy.linalg.norm(rotation)
norm_rotation = rotation / numpy.linalg.norm(rotation)
base_points = [point_x, point_y, point_z]
points = numpy.dot(base_points, norm_rotation) * norm_rotation
points_difference = base_points - points
points_transform = numpy.cross(norm_rotation, base_points)
rotated_points = points + points_difference * numpy.cos(rotation_angle) + points_transform * numpy.sin(rotation_angle)
rotated_point_x = rotated_points[0]
rotated_point_y = rotated_points[1]
return(rotated_point_x, rotated_point_y)
given an array of points my program should in theory, Find the two furthest points from each other. Then calculate the angle that those two points make with the x axis. Then in rotate all the points in the array around the averaged center of all the points by that angle. For some reason my translation function to rotate all the points around the center is not working it is giving me unexpected values. I am fairly sure the math I am using to do this is accurate since I tested the formula I am using using wolfram alpha and plotted the points on desmos. I am not sure what's wrong with my code because it keeps giving me unexpected output. Any help would greatly be appreciated.
This is the code to translate the array:
def translation(array,centerArray):
array1=array
maxDistance=0
point1=[]
point2=[]
global angle
for i in range(len(array1)):
for idx in range(len(array1)):
if(maxDistance<math.sqrt(((array1[i][0]-array1[idx][0])**2)+((array1[i][1]-array1[idx][1])**2)+((array1[i][2]-array1[idx][2])**2))):
maxDistance=math.sqrt(((array1[i][0]-array1[idx][0])**2)+((array1[i][1]-array1[idx][1])**2)+((array1[i][2]-array1[idx][2])**2))
point1 = array1[i]
point2 = array1[idx]
angle=math.atan2(point1[1]-point2[1],point1[0]-point2[0]) #gets the angle between two furthest points and xaxis
for i in range(len(array1)): #this is the problem here
array1[i][0]=((array[i][0]-centerArray[0])*math.cos(angle)-(array[i][1]-centerArray[1])*math.sin(angle))+centerArray[0] #rotate x cordiate around center of all points
array1[i][1]=((array[i][1]-centerArray[1])*math.cos(angle)+(array[i][0]-centerArray[0])*math.sin(angle))+centerArray[1] #rotate y cordiate around center of all points
return array1
This is the code I am using to test it. tortose is what I set turtle graphics name as
tortose.color("violet")
testarray=[[200,400,9],[200,-100,9]] #array of 2 3d points but don't worry about z axis it will not be used for in function translation
print("testsarray",testarray)
for i in range(len(testarray)): #graph points in testarray
tortose.setposition(testarray[i][0],testarray[i][1])
tortose.dot()
testcenter=findCenter(testarray) # array of 1 point in the center of all the points format center=[x,y,z] but again don't worry about z
print("center",testcenter)
translatedTest=translation(testarray,testcenter) # array of points after they have been translated same format and size of testarray
print("translatedarray",translatedTest) #should give the output [[-50,150,9]] as first point but instead give output of [-50,-99.999999997,9] not sure why
tortose.color("green")
for i in range(len(testarray)): #graphs rotated points
tortose.setposition(translatedTest[i][0],translatedTest[i][1])
tortose.dot()
print(angle*180/3.14) #checks to make sure angle is 90 degrees because it should be in this case this is working fine
tortose.color("red")
tortose.setposition(testcenter[0],testcenter[1])
tortose.dot()
find center code finds the center of all points in array don't worry about z axis since it is not used in translation:
def findCenter(array):
sumX = 0
sumY = 0
sumZ = 0
for i in range(len(array)):
sumX += array[i][0]
sumY += array[i][1]
sumZ += array[i][2]
centerX= sumX/len(array)
centerY= sumY/len(array)
centerZ= sumZ/len(array)
#print(centerX)
#print(centerY)
#print(centerZ)
centerArray=[centerX,centerY,centerZ]
return centerArray
import math
import turtle
tortose = turtle.Turtle()
tortose.penup()
my expected output should be a point at (-50,150) but it is giving me a point at (-50,-99.99999999999997)
This is a common mistake when doing in-place rotations:
array1[i][0]= ...
array1[i][1]= ... array[i][0] ...
First you update array1[i][0]. Then you update array1[i][1], but you use the new value when you should use the old value. Instead, temporarily store the old value:
x = array1[i][0]
array1[i][0]=((array[i][0]-centerArray[0])*math.cos(angle)-(array[i][1]-centerArray[1])*math.sin(angle))+centerArray[0] #rotate x cordiate around center of all points
array1[i][1]=((array[i][1]-centerArray[1])*math.cos(angle)+(x-centerArray[0])*math.sin(angle))+centerArray[1] #rotate y cordiate around center of all points
Please help my poor knowledge of signal processing.
I want to smoothen some data. Here is my code:
import numpy as np
from scipy.signal import butter, filtfilt
def testButterworth(nyf, x, y):
b, a = butter(4, 1.5/nyf)
fl = filtfilt(b, a, y)
return fl
if __name__ == '__main__':
positions_recorded = np.loadtxt('original_positions.txt', delimiter='\n')
number_of_points = len(positions_recorded)
end = 10
dt = end/float(number_of_points)
nyf = 0.5/dt
x = np.linspace(0, end, number_of_points)
y = positions_recorded
fl = testButterworth(nyf, x, y)
I am pretty satisfied with results except one point:
it is absolutely crucial to me that the start and end point in returned values equal to the start and end point of input. How can I introduce this restriction?
UPD 15-Dec-14 12:04:
my original data looks like this
Applying the filter and zooming into last part of the graph gives following result:
So, at the moment I just care about the last point that must be equal to original point. I try to append copy of data to the end of original list this way:
the result is as expected even worse.
Then I try to append data this way:
And the slice where one period ends and next one begins, looks like that:
To do this, you're always going to cheat somehow, since the true filter applied to the true data doesn't behave the way you require.
One of the best ways to cheat with your data is to assume it's periodic. This has the advantages that: 1) it's consistent with the data you actually have and all your changing is to append data to the region you don't know about (so assuming it's periodic as as reasonable as anything else -- although may violate some unstated or implicit assumptions); 2) the result will be consistent with your filter.
You can usually get by with this by appending copies of your data to the beginning and end of your real data, or just small pieces, depending on your filter.
Since the FFT assumes that the data is periodic anyway, that's often a quick and easy approach, and is fully accurate (whereas concatenating the data is an estimation of an infinitely periodic waveform). Here's an example of the FFT approach for a step filter.
import numpy as np
import matplotlib.pyplot as plt
x = np.arange(0, 128)
y = (np.sin(.22*(x+10))>0).astype(np.float)
# filter
y2 = np.fft.fft(y)
f0 = np.fft.fftfreq(len(x))
y2[(f0<-.25) | (f0>.25)] = 0
y3 = abs(np.fft.ifft(y2))
plt.plot(x, y)
plt.plot(x, y3)
plt.xlim(-10, 140)
plt.ylim(-.1, 1.1)
plt.show()
Note how the end points bend towards each other at either end, even though this is not consistent with the periodicity of the waveform (since the segments at either end are very truncated). This can also be seen by adjusting waveform so that the ends are the same (here I used x+30 instead of x+10, and here the ends don't need to bend to match-up so they stay at level with the end of the data.
Note, also, to have the endpoints actually be exactly equal you would have to extend this plot by one point (at either end), since it periodic with exactly the wavelength of the original waveform. Doing this is not ad hoc though, and the result will be entirely consistent with your analysis, but just representing one extra point of what was assumed to be infinite repeats all along.
Finally, this FFT trick works best with waveforms of length 2n. Other lengths may be zero padded in the FFT. In this case, just doing concatenations to either end as I mentioned at first might be the best way to go.
The question is how to filter data and require that the left endpoint of the filtered result matches the left endpoint of the data, and same for the right endpoint. (That is, in general, the filtered result should be close to most of the data points, but not necessarily exactly match any of them, but what if you need a match at both endpoints?)
To make the filtered result exactly match the endpoints of a curve, one could add a padding of points at either end of the curve and adjust the y-position of this padding so that the endpoints of the valid part of the filter exactly matched the end points of the original data (without the padding).
In general, this can be done by either iterating towards a solution, adjusting the padding y-position until the ends line up, or by calculating a few values and then interpolating to determine the y-positions that would be required for the matched endpoints. I'll do the second approach.
Here's the code I used, where I simulated the data as a sine wave with two flat pieces on either side (note, that these flat pieces are not the padding, but I'm just trying to make data that looks a bit like the OPs).
import numpy as np
from scipy.signal import butter, filtfilt
import matplotlib.pyplot as plt
#### op's code
def testButterworth(nyf, x, y):
#b, a = butter(4, 1.5/nyf)
b, a = butter(4, 1.5/nyf)
fl = filtfilt(b, a, y)
return fl
def do_fit(data):
positions_recorded = data
#positions_recorded = np.loadtxt('original_positions.txt', delimiter='\n')
number_of_points = len(positions_recorded)
end = 10
dt = end/float(number_of_points)
nyf = 0.5/dt
x = np.linspace(0, end, number_of_points)
y = positions_recorded
fx = testButterworth(nyf, x, y)
return fx
### simulate some data (op should have done this too!)
def sim_data():
t = np.linspace(.1*np.pi, (2.-.1)*np.pi, 100)
y = np.sin(t)
c = np.ones(10, dtype=np.float)
z = np.concatenate((c*y[0], y, c*y[-1]))
return z
### code to find the required offset padding
def fit_with_pads(v, data, n=1):
c = np.ones(n, dtype=np.float)
z = np.concatenate((c*v[0], data, c*v[1]))
fx = do_fit(z)
return fx
def get_errors(data, fx):
n = (len(fx)-len(data))//2
return np.array((fx[n]-data[0], fx[-n]-data[-1]))
def vary_padding(data, span=.005, n=100):
errors = np.zeros((4, n)) # Lpad, Rpad, Lerror, Rerror
offsets = np.linspace(-span, span, n)
for i in range(n):
vL, vR = data[0]+offsets[i], data[-1]+offsets[i]
fx = fit_with_pads((vL, vR), data, n=1)
errs = get_errors(data, fx)
errors[:,i] = np.array((vL, vR, errs[0], errs[1]))
return errors
if __name__ == '__main__':
data = sim_data()
fx = do_fit(data)
errors = vary_padding(data)
plt.plot(errors[0], errors[2], 'x-')
plt.plot(errors[1], errors[3], 'o-')
oR = -0.30958
oL = 0.30887
fp = fit_with_pads((oL, oR), data, n=1)[1:-1]
plt.figure()
plt.plot(data, 'b')
plt.plot(fx, 'g')
plt.plot(fp, 'r')
plt.show()
Here, for the padding I only used a single point on either side (n=1). Then I calculate the error for a range of values shifting the padding up and down from the first and last data points.
For the plots:
First I plot the offset vs error (between the fit and the desired data value). To find the offset to use, I just zoomed in on the two lines to find the x-value of the y zero crossing, but to do this more accurately, one could calculate the zero crossing from this data:
Here's the plot of the original "data", the fit (green) and the adjusted fit (red):
and zoomed in the RHS:
The important point here is that the red (adjusted fit) and blue (original data) endpoints match, even though the pure fit doesn't.
Is this a valid approach? Of the various options, this seems the most reasonable since one isn't usually making any claims about the data that isn't being shown, and also for show region has an accurately applied filter. For example, FFTs usually assume the data is zero or periodic beyond the boundaries. Certainly, though, to be precise one should explain what was done.
I'm trying to fill in a structured grid with an analytical field, but despite reading the vtk docs, I haven't found out how to actually set scalar values at the grid points or the set the spacing/origin info of the grid. Starting from the code below, how do I
associate spatial information with the grid (ie cell 0,0,0 is at coordinates 0,0,0, the spacing is dx in every direction)
associate scalar values with each grid point. To start, I just need one, but eventually I'd like to store 3 pieces of data at each point (not a vector, 3 distinct scalars).
grid = vtk.vtkStructuredGrid()
numPoints = int((maxGrid - minGrid)/dx)
grid.SetDimensions(numPoints, numPoints, numPoints)
In VTK there are 3 types of "structured" grids, vtkImageData (vtkUniformGrid derives from this), vtkRectilinearGrid, and vtkStructuredGrid. They are all structured in the sense that the topology is set. vtkImageData has constant spacing between points and is axis aligned, vtkRectilinearGrid is axis aligned but can vary the spacing in each axis direction, and vtkStructuredGrid has arbitrarily located points (cells may not be valid though).
For what you want to do you should do:
from vtk import *
dx = 2.0
grid = vtkImageData()
grid.SetOrigin(0, 0, 0) # default values
grid.SetSpacing(dx, dx, dx)
grid.SetDimensions(5, 8, 10) # number of points in each direction
# print grid.GetNumberOfPoints()
# print grid.GetNumberOfCells()
array = vtkDoubleArray()
array.SetNumberOfComponents(1) # this is 3 for a vector
array.SetNumberOfTuples(grid.GetNumberOfPoints())
for i in range(grid.GetNumberOfPoints()):
array.SetValue(i, 1)
grid.GetPointData().AddArray(array)
# print grid.GetPointData().GetNumberOfArrays()
array.SetName("unit array")