Find peaks in a matrix containing zeros

Dear All Community Members,

I have two matrices, 'A' and 'B'. Matrix 'A' represents the time history of velocities, while matrix 'B' represents the corresponding times. I want to extract all the peak velocity values from matrix 'A' that are above a threshold value of 3.

Matrix A is as follows:

A=[ 1 3 -1 1 5 -1 0;1 6 3 -2 0 0 0;2 3 9 -2 1 11 -1;4 1 -2 8 5 0 0;2 -2 6 -1 0 0 0;-1 13 -2 0 0 0 0];

Matrix B is as follows:

B=[1 1.5 2.5 3 3.5 4 0;1.5 3 5 6 0 0 0;0.5 2 2.5 2.75 3 4 5;2 4 4.25 4.5 6 0 0;1 2 3 4 0 0 0;0.5 3 4 0 0 0 0];

Specifically, I am looking to extract the value 5 from the first row, the value 6 from the second row, the values 9 and 11 from the third row, and then values 8,6,13 from the from the fourth and to the last row in matrix A.

I have tried the findpeaks function, but I am getting errors because of the ending zeros in matrix B.

Can anyone help me out?

Thank you in advance.

[Dyuman Joshi]

I am not sure if I get the logic used to extract the values.

In the 4th row, 4 is above 3, but they are not extracted.

How does the values from B come into play here?

Answer 1

[Stephen23]

A = [1,3,-1,1,5,-1,0;1,6,3,-2,0,0,0;2,3,9,-2,1,11,-1;4,1,-2,8,5,0,0;2,-2,6,-1,0,0,0;-1,13,-2,0,0,0,0]

A = 6×7

1 3 -1 1 5 -1 0 1 6 3 -2 0 0 0 2 3 9 -2 1 11 -1 4 1 -2 8 5 0 0 2 -2 6 -1 0 0 0 -1 13 -2 0 0 0 0

F = @(a)findpeaks(a,'Threshold',3);
C = cellfun(F,num2cell(A,2),'uni',0)

C = 6×1 cell array
    {[   5]}
    {[   6]}
    {[9 11]}
    {[   8]}
    {[   6]}
    {[  13]}

[Dyuman Joshi]

Not a comment on the answer, but on MATLAB.

I find it weird that islocalmax() considers 4 as a local maxima, but findpeaks() with a threshold of 3 does not consider it as a local maxima.

I am aware that the findpeaks() is defined as such, but then the labelling that it finds the local maxima is plainly wrong and 'maxima' should be removed from the documentation.

A = [1,3,-1,1,5,-1,0;1,6,3,-2,0,0,0;2,3,9,-2,1,11,-1;4,1,-2,8,5,0,0;2,-2,6,-1,0,0,0;-1,13,-2,0,0,0,0]

A = 6×7

1 3 -1 1 5 -1 0 1 6 3 -2 0 0 0 2 3 9 -2 1 11 -1 4 1 -2 8 5 0 0 2 -2 6 -1 0 0 0 -1 13 -2 0 0 0 0

islocalmax(A)

ans = 6×7 logical array
   0   0   0   0   0   0   0
   0   1   0   0   0   0   0
   0   0   1   0   0   1   0
   1   0   0   1   1   0   0
   0   0   1   0   0   0   0
   0   0   0   0   0   0   0

[Stephen23]

V = [1,1,2,4,2,-1]

V = 1×6

1 1 2 4 2 -1

X = findpeaks(V)

X = 4

Y = islocalmax(V)

Y = 1×6 logical array
   0   0   0   1   0   0

By specifying the threshold explicitly then of course I can change what peaks are detected:

X = findpeaks(V, 'Threshold',3)

X =

  1×0 empty double row vector

Clearly 4 is not >=3 compared to the 2's on either side, so it is (according to my requirement) not a maxima. In fact, there are no maxima in that vector according to the threshold that I specified.

"but then the labelling that it finds the local maxima is plainly wrong and 'maxima' should be removed from the documentation."

I do not follow your argument: the definition of what is a local maxima depends on the prominence, the window, etc... I do not see what is "plainly" wrong here.

[Askeladden2]

Thank you for your quick response!