Is there a way to crop an array depending on the values that I want through a function as the values change?

I currently have a graph that has multiple flat maxima but I only want the index values of the last one. I am currently using islocalmax(data,'flatselection','all') to get all of the local maxima that exist and that then gives me a logical array which I can then either get the index values of all the 1's but I then want the last collection of 1's or the last collection of consecutive indexes. I was just wondering if there was a way to specifically select these values? I could manually find the start point of the last set and then set it to the end but as I want to use it on multiple data sets I want it to be automatic.

peaks = islocalmax(test.CAN_EMTrq,'FlatSelection','all');

peaks1 = find(peaks==1);

This is the code I am using right now.

The table either looks like: 000011111000100001111101111 and I want the last 4 1's or it looks like: 12,13,14,15,34,35,36,37,38,56,57,58,59 and I want the last 4 numbers.

Thanks

Answer 1

[Mathieu NOE]

hello

try this :

a = [0 0 0 0 1 1 1 1 1 0 0 0 1 0 0 0 0 1 1 1 1 1 0 1 1 1 1];
ind = logical(a);

[begin,ends] = find_start_end_group(ind);
duration = ends - begin;
% last group of duration >=3 consecutive samples 
ind = duration>=3;
begin = begin(ind);
ends = ends(ind);

% start and end indexes of last buffer of  1's
begin = begin(end)

begin = 24

ends = ends(end)

ends = 27

function [begin,ends] = find_start_end_group(ind)
% This locates the beginning /ending points of data groups
% Important : ind must be a LOGICAL array

    D = diff([0;ind(:);0]);
    begin = find(D == 1);
    ends = find(D == -1) - 1;
end

Answer 2

[Image Analyst]

The binary option is trivial if you have the Image Processing Toolbox

signal = logical([0,0,0,0,1,1,1,1,1,0,0,0,1,0,0,0,0,1,1,1,1,1,0,1,1,1,1]);
labeledSignal = bwlabel(signal)

labeledSignal = 1x27

0 0 0 0 1 1 1 1 1 0 0 0 2 0 0 0 0 3 3 3 3 3 0 4 4 4 4

<mw-icon class=""></mw-icon>

<mw-icon class=""></mw-icon>

(the display is being truncated and the 4's aren't showing up but they're there) but I don't know exactly what you mean when you say you want the last 4 1's. What does that mean exactly? Do you want to extract a vector of [1 1 1 1] or want their indexes or what????

for the floating point case you need to find out how flat is flat because it's a judgment call. You might want to call some kind of clustering method like dbscan or something. See

https://www.mathworks.com/help/stats/cluster-analysis.html

Machine Learning, classification, clustering

Cluster analysis - Wikipedia, the free encyclopedia

Once you've collected the data into clusters it will be easy to extract the last cluster. See this demo:

v = [12,13,14,15,34,35,36,37,38,56,57,58,59]

v = 1x13

12 13 14 15 34 35 36 37 38 56 57 58 59

<mw-icon class=""></mw-icon>

<mw-icon class=""></mw-icon>

x = 1 : numel(v);
data = [v; x]'

data = 13x2

12 1 13 2 14 3 15 4 34 5 35 6 36 7 37 8 38 9 56 10

<mw-icon class=""></mw-icon>

<mw-icon class=""></mw-icon>

scatter(x, v);

result = dbscan(data,5,2)

result = 13x1

1 1 1 1 2 2 2 2 2 3

<mw-icon class=""></mw-icon>

<mw-icon class=""></mw-icon>

gscatter(x, v, result); % Visualize the groups it found

% Get last group
indexes = result == max(result)

indexes = 13x1 logical array
   0
   0
   0
   0
   0
   0
   0
   0
   0
   1
   1
   1
   1

lastDataGroup = v(indexes)

lastDataGroup = 1x4

56 57 58 59

<mw-icon class=""></mw-icon>

<mw-icon class=""></mw-icon>