How to count the total number of blink?

2 views (last 30 days)
Hi, I'm doing research on how count total number of blinks, is there any good resource for this task?
Thank You

Accepted Answer

Bora Eryilmaz
Bora Eryilmaz on 21 Dec 2022
Edited: Bora Eryilmaz on 21 Dec 2022
You can use a peak detection algorithm such as the islocalmax() command: https://www.mathworks.com/help/matlab/ref/islocalmax.html
x = randn(100,1);
plot(x)
hold on
level = 1.0;
yline(level, 'r')
% Only find peaks above the level.
I = (x < level);
x(I) = 0;
% Indices of all peaks above level.
J = islocalmax(x);
count = sum(J) % Number of peaks
count = 17
plot(find(J), x(J), 'ro')
% Indices of the peaks above level that are at least 5 x-units apart.
J = islocalmax(x, 'MinSeparation', 5);
count = sum(J) % Number of peaks
count = 11
plot(find(J), x(J), 'bx')
hold off
  5 Comments

Sign in to comment.

More Answers (1)

Image Analyst
Image Analyst on 21 Dec 2022
Not sure what "detect" means to you. But if the red line is your threshold and signals above the threshold are a blink, then you can count them with bwlabel". If you want to know what indexes contains a blink signal, you can use regionprops.
signal = rand(1, 200); % Sample signal
plot(signal, 'b-');
grid on;
xlabel('Index');
ylabel('Signal');
threshold = 0.9;
yline(threshold, 'Color', 'r', 'LineWidth',2)
binarySignal = signal > threshold;
[~, numBlinks] = bwlabel(binarySignal);
% If you want indexes of the blink runs, then you can use regionprops
props = regionprops(binarySignal, 'PixelIdxList')
props = 12×1 struct array with fields:
PixelIdxList
% Print them out
for k = 1 : numel(props)
theseIndexes = props(k).PixelIdxList;
fprintf('For region #%d, it has indexes : ', k)
fprintf('%d ', theseIndexes)
fprintf('\n')
end
For region #1, it has indexes :
13
For region #2, it has indexes :
61
For region #3, it has indexes :
71
For region #4, it has indexes :
75
For region #5, it has indexes :
81
For region #6, it has indexes :
84 85
For region #7, it has indexes :
100
For region #8, it has indexes :
140 141
For region #9, it has indexes :
148 149
For region #10, it has indexes :
174
For region #11, it has indexes :
177
For region #12, it has indexes :
194
  4 Comments
Bora Eryilmaz
Bora Eryilmaz on 21 Dec 2022
Edited: Bora Eryilmaz on 21 Dec 2022
The code in this answer would require the Image Processing Toolbox.
Image Analyst
Image Analyst on 21 Dec 2022
Uh, that's not my code. You should contact the author for a fix.

Sign in to comment.

Community Treasure Hunt

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

Start Hunting!