# Remove consecutive duplicates from matrix

3 views (last 30 days)
Simon Meyer on 26 Nov 2020
Commented: Ameer Hamza on 30 Nov 2020
Hello,
i need to remove duplicates form my matrix, but only if they are consecutive duplicates.
Lets suppose i have the following list of coordinates in a matrix, with for example x, y and z coordinates, that is describing a path.
A=[ 1, 0, 3
2, 1, 3
2, 1, 3
2, 2, 3
1, 0, 3]
Now, you can see that the point [2, 1, 3] is a consecutive duplicate, that I want to remove (because it's unuseful information).
But I do not want to remove the (duplicate) row at the end, [1, 0, 3], even if this row is already there in the beginning.
Also, the sorting shouldn't be changed. The desired output should be
A=[ 1, 0, 3
2, 1, 3
2, 2, 3
1, 0, 3]
I cannot use 'unique' for that, because that would filter also the last line.
What is the smartest way to do that?
Thank you very much for your help!

Ameer Hamza on 26 Nov 2020
Try this
A=[ 1, 0, 3
2, 1, 3
2, 1, 3
2, 2, 3
1, 0, 3];
idx = find(~any(diff(A), 2))+1;
A(idx, :) = [];

Show 1 older comment
Ameer Hamza on 26 Nov 2020
A=[ 1, 0, 3
2, 1, 3
2, 1, 3
2, 2, 3
1, 0, 3];
First apply the diff() on each column. It calculates difference of consecutive elements
>> diff(A)
ans =
1 1 0
0 0 0
0 1 0
-1 -2 0
Now I want the rows in which all elements are zero
>> ~any(diff(A), 2)
ans =
4×1 logical array
0
1
0
0
the number of rows are one less then the number of rows in A. Since first row can never be removed so I expand the vector like this
>> idx = [false; ~any(diff(A), 2)]
idx =
5×1 logical array
0
0
1
0
0
It shows the rows I want to remove. So I just remove them
>> A(idx, :) = []
A =
1 0 3
2 1 3
2 2 3
1 0 3
Simon Meyer on 30 Nov 2020
Thank you very much! :)
Ameer Hamza on 30 Nov 2020
I am glad to be of help!

R2018b

### Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!