I have two sets of data points; effectively, one is from a preimage and the other from its image, but I do not know the rule between the two. This rule/function is nonlinear.
I've collected many data points of corresponding locations on both images, and I was wondering if anyone knew of a way to find a more complete mapping. That is, does anyone know the best way to find a mapping from R^2 to R^2 with an extensive set of sample points. This mapping is one-to-one and onto.
My goal is to use the data I've found to find a polynomial function that takes in some x,y coordinate from the preimage, and outputs the shifted coordinates.
edit: I have sample points along the domain and their corresponding points in the image, but not for every point in the domain. I want to be able to input any point (only integer values) in the domain and output the shifted point.
I don't think polynomial is easy (or easy to guarantee is a bijection). The obvious thing to do is to
Construct the delaunay triangulation of the known points in the domain.
For each delaunay triangle the mapping is just the linear mapping which interpolates the map on the vertices.
Then, when you have a random point, look up its delaunay triangle, and apply the requisite map.
I believe that all of the above can be done via scipy.spatial.delaunay.
The transformation you're trying to find sounds a lot like what's accomplished in Geographic Information Systems using a technique called rubber-sheeting https://en.wikipedia.org/wiki/Rubbersheeting
Igor Rivin's description of a process using a Delaunay triangulation is pretty much the solution that's used in such systems. Some systems will use a Barycentric coordinate system rather than a linear mapping to try to reduce the appearance of triangle-related artifacts in the transformed image.
What you are describing also sounds a bit like the "morphing" special effect used in video. Maybe a web search on that topic would turn up some leads for you.
Related
I'm a wee bit stuck.
I have a 3D point cloud (an array of (n,3) vertices), in which I am trying to generate a 3D triangular mesh from. So far I have had no luck.
The format my data comes in:
(x,y) values in regularly spaced (z) intervals. Think of the data as closed loop planar contours stored slice by slice in the z direction.
The vertices in my data must be absolute positions for the mesh triangles (i.e. I don't want them to be smoothed out such that the volume begins to change shape, but linear interpolation between the layers is fine).
Illustration:
Z=2. : ..x-------x... <- Contour 2
Z=1.5: ...\......|... <- Join the two contours into a mesh.
Z=1. : .....x----x... <- Contour 1
Repeat for n slices, end up with an enclosed 3D triangular mesh.
Things I have tried:
Using Open3D:
The rolling ball (pivot) method can only get 75% of the mesh completed and leaves large areas incomplete (despite a range of ball sizes). It has particular problems at the top and bottom slices where there tends to be large gaps in the middle (i.e. a flat face).
The Poisson reconstruction method smooths out the volume too much and I no longer have an accurate representation of the volume. This occurs at all depths from 3-12.
CGAL:
I cannot get this to work for the life of me. SWIG is not very good, the implementation of CGAL using SWIG is also not very good.
There are two PyBind implementations of CGAL however they have not incorporated the 3D triangulation libraries from CGAL.
Explored other modules like PyMesh, TriMesh, TetGen, Scikit-Geometry, Shapely etc. etc. I may have missed the answer somewhere along the line.
Given that my data is a list of closed-loop planar contours, it seems as though there must be some simple solution to just "joining" adjacent slice contours into one big 3d mesh. Kind of like you would in blender.
There are non-python solutions (like MeshLab) that may well solve these problems, but I require a python solution. Does anyone have any ideas? I've had a bit of a look into VTK and ITK but haven't found exactly what I'm looking for as of yet.
I'm also starting to consider that maybe I can interpolate intermediate contours between slices, and fill the contours on the top and bottom with vertices to make the data a bit more "pivot ball" method friendly.
Thank you in advance for any help, it is appreciated.
If there is a good way of doing this that isn't coded yet, I promise to code it and make it available for people in my situation :)
Actually there are two ways of having meshlab functionality in python:
The first is MeshLabXML (https://github.com/3DLIRIOUS/MeshLabXML ) a third party, is a Python scripting interface to meshlab scripting interface
the second is PyMeshLab (https://github.com/cnr-isti-vclab/PyMeshLab ) an ongoing effort done by the MeshLab authors, (currently in alpha stage) to have a direct Python bindings to all the meshlab filters
There is a very neat paper titled "Technical Note: an algorithm and software for conversion of radiotherapy contour‐sequence data to ready‐to‐print 3D structures" in the Journal of Medical Physics that describes this problem quite nicely. No python packages are required, however it is more easily implemented with numpy. No need for any 3D packages.
A useful excerpt is provided:
...
The number of slices (2D contours) constituting the specified structure is determined.
The number of points in each slice is determined.
Cartesian coordinates of each of the points in each slice are extracted and stored within dedicated data structures...
Numbers of points in each slice (curve) are re‐arranged in such a way, that the starting points (points with indices 0) are the closest points between the subsequent slices. Renumeration starts at point 0, slice 0 (slice with the lowest z coordinate).
Orientation (i.e., the direction determined by the increasing indices of points with relation to the interior/exterior of the curve) of each curve is determined. If differences between slices are found, numbering of points in non‐matching curves (and thus, orientation) is reversed.
The lateral surface of the considered structure is discretized. Points at the neighboring layers are arranged into threes, constituting triangular facets for the STL file. For each triangle the closest points with the subsequent indices from each layer are connected.
Lower and upper base surfaces of the considered structure are discretized. The program iterates over every subsequent three points on the curve and checks if they belong to a convex part of the edge. If yes, they are connected into a facet, and the middle point is removed from further iterations.
So basically it's a problem of aligning datasets in each slice to the nearest value of each slice. Then aligning the orientation of each contour. Then joining the points between two layers based on distance.
The paper also provides code to do this (for a DICOM file), however I re-wrote it myself and it works a charm.
I hope this helps others! Make sure you credit the author's in any work you do that uses this.
A recent feature of pymadcad can do things like this, not sure through if it fits your exact expectation in term of "pivot ball" or such things, checkout the doc for blending
Starting from a list of outlines, it can generate blended surfaces to join them:
For your purpose, I guess the best is one of:
blendpair(line1, line2)
junction(*lines)
I have unstructured (taken in no regular order) point cloud data (x,y,z) for a surface. This surface has bulges (+z) and depressions (-z) scattered around in an irregular fashion. I would like to generate some surface that is a function of the original data points and then be able to input a specific (x,y) and get the surface roughness value from it (z value). How would I go about doing this?
I've looked at scipy's interpolation functions, but I don't know if creating a single function for the entire surface is the correct approach? Is there a technical name for what I am trying to do? I would appreciate any suggestions/direction.
I don't know if creating a single function for the entire surface is the correct approach?
I guess this depends on your data. Let's assume the base form of your surface is spherical. Then you can model it as such.
If your surface is more complex then a sphere you might can still model the neighborhood of (x,y) as such. Maybe you could even consider your surface as plain in the near neighborhood of (x,y).
What you are trying to do, can be called surface fitting, or two-dimensional curve fitting. You would be able to find lots of available algorithms by searching for those terms. Now, the choice of the particular algorithm/method should be dictated:
by the origin of your data (there are specialized algorithms or variations of more common ones that are tailored for certain application areas)
by the future use of your data (depending on what you are going to do with it, maybe you need to be able to calculate derivatives easily, etc)
It is not easy to represent complicated data (especially the noisy one) using a single function. Thus there is a lot of research about it. However, in a lot of applications curve-fitting is very successful and very widely used.
I have two coordinate systems for each record in my dataset. Lat-lon coordinates and what I suppose is utm x-y coordinates.
50% of my dataset only has x-y data without lat-lon, viceversa is 6%.
There is a good portion of the dataset that has both (33%) for each single record.
I wanted to know if there is a way to take advantage of the intersection (and maybe the x-y only part, since it's the biggest) to obtain a full dataset with only one coordinate system that makes sense. The problem is that after a little bit of preprocessing, they look "relaxed" in a different way and the intersection doesn't really match. The scatter plot shows what I believe to be a non linear, warped relationship from one system of coordinates to another. With this, I mean that normalizing them both to [0;1] and centering them to (0,0) (by subtracting the mean), gives two slightly different point distributions, and apparently a coefficient multiplication to scale one down to look like the other is not enough to get them to match completely. Looks like some more complicated relationship between the two.
I also tried to use an external library called utm to convert the lat-long coordinates to x-y to have a third pair of attributes (let's call it my_xy), only to find out that is not matching even one of the first two systems, instead it shows another slight warp.
Notes: When I say I do not have data from one coordinate system, assume NaN.
Furthermore, I know the warping could be a result of the fundamental geometrical differences between latlon and xy systems, but I still do not know what else I could try, if the utm conversion and the scaling did not work.
Blue: latlon, Red: original xy, Green: my_xy calculated from latlon
I have to images, one simulation, one real data, with bright spots.
Simulation:
Reality:
I can detect the spots just fine and get the coordinates. Now I need to compute transformation matrix (scale, rotation, translation, maybe shear) between the two coordinate systems. If needed, I can pick some (5-10) corresponding points by hand to give to the algorithm
I tried a lot of approaches already, including:
2 implementations of ICP:
https://engineering.purdue.edu/kak/distICP/ICP-2.0.html#ICP
https://github.com/KojiKobayashi/iterative_closest_point_2d
Implementing affine transformations:
https://math.stackexchange.com/questions/222113/given-3-points-of-a-rigid-body-in-space-how-do-i-find-the-corresponding-orienta/222170#222170
Implementations of affine transformations:
Determining a homogeneous affine transformation matrix from six points in 3D using Python
how to perform coordinates affine transformation using python? part 2
Most of them simply fail somehow like this:
The red points are the spots from the simulation transformed into the reality - coordinate system.
The best approach so far is this one how to perform coordinates affine transformation using python? part 2 yielding this:
As you see, the scaling and translating mostly works, but the image still needs to be rotated / mirrored.
Any ideas on how to get a working algorithm? If neccessary, I can provide my current non-working implementations, but they are basically as linked.
I found the error.
I used plt.imshow to display both the simulated and real image and from there, pick the reference points from which to calculate the transformation.
Turns out, due to the usual array-to-image-index-flipping-voodoo (or a bad missunderstanding of the transformation on my side), I need to switch the x and y indices of the reference points from the simulated image.
With this, everything works fine using this how to perform coordinates affine transformation using python? part 2
I would like to find the rotation and location for a polygon that maximizes how large it can be scaled up within the constraints of fitting within a larger polygon.
Current idea is to use scipy optimization routines for optimizing position and rotation parameters to maximize the scaling parameter, and shapely to add a constraint that the polygon is contained. This seems like it'll be slow and not particularly elegant.
Other ideas?
This problem sounds like it might be NP-Hard. Given a candidate solution you can't really be certain that it is the best solution. It seems like you'd need to try to use some sort of incremental randomized search.
If the internal polygon is maximally scaled, then there are at least 4 pairs "internal vertice - external edge" or "external vertice - internal edge" where the vertice is on the edge.
Let's take all 4s of vertice-edge pairs. For every one we get the system of linear equations for two reference points' coordinates. If it has one solution, we verify that there are no intersections and if OK we remember the coordinates and size of the internal polygon.
This is an exact solution, but it's slow. On the other hand, scipy optimization routines are likely to find a local maximum which is not the global one.