delete Xn,Yn and Zn points from point cloud that are equal to Xn, Yn and Zn point from another cloud

Hi all,

I got X,Y and Z point cloud, (4500x3 matrix) after plotting this figure I used the brushing tool to select some point of the data wich i saved in a new varriable. (600x3 matrix --> also x, y and z coördinates)

Now I want to delete the selected data from the original point cloud (4500x3 matrix) - (600x3 matrix) --> (3900x 3 matrix) so that al the data were for the (X & Y & Z) from the Original pointcloud = (x & y & z) of the selected point cloud are deleted.

Thanks in advance!

Answer 1

[Baltam]

Suppose your variables are called

OriginalPointCloud % (4500x3)

BrushData % (600x3)

NewPointCloud = OriginalPointCloud;
NewPointCloud(ismember(OriginalPointCloud,BrushData,'rows')) = NaN;

Kind regards, Baltam

[Allan Uytterhoeven]

Hello Baltman,

first of all thanks for your help! This already helps a lot but still got a problem solving the problem. For example:

>> OriginalPointCloud= [ 1 2 3; 1 2 4; 5 4 8; 2 2 2]

OriginalPointCloud =

     1     2     3
     1     2     4
     5     4     8
     2     2     2

>> BrushData= [ 5 4 8; 1 2 4]

BrushData =

     5     4     8
     1     2     4

>> NewPointCloud=OriginalPointCloud

NewPointCloud =

     1     2     3
     1     2     4
     5     4     8
     2     2     2

>> NewPointCloud(ismember(OriginalPointCloud,BrushData,'rows')=NaN

NewPointCloud(ismember(OriginalPointCloud,BrushData,'rows')=NaN | Error: The expression to the left of the equals sign is not a valid target for an assignment.

At the end i get an error.

But than i tried the following:

>> Lia= ismember(OriginalPointCloud,BrushData,'rows')

Lia =

     0
     1
     1
     0

Maybe u know some kind of trick with this last row vector?

Kind regards, Allan Uytterhoeven

[Baltam]

I'm sorry, I forgot something.

OriginalPointCloud= [ 1 2 3; 1 2 4; 5 4 8; 2 2 2];
BrushData= [ 5 4 8; 1 2 4];
NewPointCloud=OriginalPointCloud;
NewPointCloud(ismember(OriginalPointCloud,BrushData,'rows'),:) = NaN

Note the ":" added to the second time we call for NewPointCloud. This is needed because we do operations on the rows but to assign the NaNs we must not forget the columns. This should work.

[Allan Uytterhoeven]

Just found the perfect solution!! >> A= [1 1 1; 2 2 2; 1 2 3; 2 5 6; 8 4 5]

A =

     1     1     1
     2     2     2
     1     2     3
     2     5     6
     8     4     5

>> B= [ 1 1 1; 8 4 5]

B =

     1     1     1
     8     4     5

>> C=setdiff(A,B,'rows')

C =

     1     2     3
     2     2     2
     2     5     6

[Baltam]

Nice, that's even better.

Good job,

Baltam