Identifying time for prolong changes in a sequence

2 views (last 30 days)
Hello
Please I have a sequence for instance
010001100000111110000100011001
And I want to identify the ones that are more than 2(onces) and the time
for this example is 5 seconds
Also another example is when I have more than one occurances of 1 greater than 2
000001110001111001000110001
Then the time are [ 3sec 4sec]
I have used the code A = seconds(x)
It only shows me a total number of 1 in seconds.
Is there a way I can get about this.
Thanks for your help in advance

Accepted Answer

Image Analyst
Image Analyst on 6 Jul 2021
Try this (requires the Image Processing Toolbox):
binarySequence = [0 1 0 0 0 1 1 0 0 0 0 0 1 1 1 1 1 0 0 0 0 1 0 0 0 1 1 0 0 1];
props = regionprops(logical(binarySequence), 'Area');
allLengths = [props.Area]
% Count number of regions longer than some number:
maxLength = max(allLengths)
for k = 1 : maxLength
longerThanN = sum(allLengths == k);
fprintf('There are %d sequences of %d.\n', longerThanN, k);
end
You get:
There are 3 sequences of 1.
There are 2 sequences of 2.
There are 0 sequences of 3.
There are 0 sequences of 4.
There are 1 sequences of 5.
  10 Comments
Tino
Tino on 12 Jul 2021
Hi Image analyst.
If I wanted only onces to show (and there group occurences in the sequence) in the result instead of both 1 and zero
for instance if I have 1111100001111000000111111
I will like the results to be
There are 1 sequences of 4 = second occurences
There are 2 sequences of 5 = first and third occurences
I will appreciate it if you answer the question for me as you ve done in the past
Best wishes
Image Analyst
Image Analyst on 12 Jul 2021
Try this:
binarySequence = [0 1 0 0 0 1 1 0 0 0 0 0 1 1 1 1 1 0 0 0 0 1 0 0 0 1 1 0 0 1]
% Measure the data
props = regionprops(logical(binarySequence), 'Area', 'PixelIdxList');
allLengths = [props.Area]
% Count number of regions longer than some number:
maxLength = max(allLengths)
for k = 1 : maxLength
% Get the labels for this length
labels = allLengths == k;
longerThanN = sum(labels);
fprintf('There are %d sequences of %d.\n', longerThanN, k);
indexes{k} = find(labels);
for k2 = 1 : length(indexes{k})
fprintf(' Sequence %d\n', indexes{k}(k2));
end
% Get 1's wherever the run has a length of k.
locations = zeros(size(binarySequence));
for k2 = 1 : length(labels)
if labels(k2)
locations(props(k2).PixelIdxList) = 1;
end
end
% Report to the command window:
%locations
end
There are 3 sequences of 1.
Sequence 1
Sequence 4
Sequence 6
There are 2 sequences of 2.
Sequence 2
Sequence 5
There are 0 sequences of 3.
There are 0 sequences of 4.
There are 1 sequences of 5.
Sequence 3

Sign in to comment.

More Answers (0)

Categories

Find more on Labels and Annotations in Help Center and File Exchange

Products


Release

R2020a

Community Treasure Hunt

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

Start Hunting!