Find the maxima in a graph
Ältere Kommentare anzeigen
I know that this question has been answered a lot of times in this forum but I am a beginner in MATLAB and I was not able to find a suitable solution for myself.
here is the link for uploaded graph. The variable which I am plotting has dimensions of 2705*196. How can i find the maximum peak of the graph.
1 Kommentar
Ricky
am 23 Mai 2013
Antworten (3)
Image Analyst
am 22 Mai 2013
Try this:
[maxDataValue, indexAtMax] = max(yourData);
3 Kommentare
Image Analyst
am 22 Mai 2013
The index value will be correct. If you say it's not, it is probably because your x axis is not in elements but in some kind of real world units. In that case you'll need a calibration factor that converts the element into real world units. You can upload your data here: http://www.mathworks.com/matlabcentral/answers/7924-where-can-i-upload-images-and-files-for-use-on-matlab-answers
Ricky
am 22 Mai 2013
Iain
am 22 Mai 2013
0 Stimmen
[m K] = max(Variable(:)) tells you the maximum value anywhere in Variable (m), and it's index (K).
4 Kommentare
Ricky
am 22 Mai 2013
Iain
am 22 Mai 2013
The index value is correct when used as: Variable(K)
Image Analyst's approach is also correct, but assumes you want the max of each column, and his indexAtMax is also correct, when used as Variable(indexAtMax(column_number),column_number)
Image Analyst
am 22 Mai 2013
Bearbeitet: Image Analyst
am 23 Mai 2013
He showed one dimensional data in his screenshot, not 2D data. That's why I didn't use (:). Plus the values will be the same no matter if the 1D data is a row vector or a column vector. A row vector will not give you the value of each column back - it will give you the max like you expect. If someone has a 2D matrix and want the overall matrix then (:) is correct. Of course it will also work with 1D arrays as well though it's not necessary.
Iain
am 23 Mai 2013
Ah - Due to the firewall I'm stuck with here, I can't see the screenshot.
José-Luis
am 22 Mai 2013
Sounds like you want the sub indexes (row, column) not the linear indexes.
your_data = rand(10,20);
[row col] = ind2sub(size(your_data),find(your_data == max(your_data(:))))
8 Kommentare
Image Analyst
am 22 Mai 2013
Ricky's data looks to me like 1D data. What does it look like to you?
José-Luis
am 22 Mai 2013
Sure does. But he provides some conflicting information. Rick says the variable has size (2075,196). My guess was the plot is not very informative, and that he meant to use surf() or the likes.
Oh well, if all questions were cristal clear, I guess it would take away some of the fun. It would make answering questions more efficient though, removing the guessing part.
Image Analyst
am 23 Mai 2013
You're right. I didn't notice that. That certainly does contradict what his plot shows.
Ricky
am 23 Mai 2013
Could you explain why the alternatives proposed here do not fit the bill? The data and the plot don't explain things, at least no to me. What do you mean by maxima? Image Analyst answer gives you the maximum for every column of data. Iain and me give you the maximum for the entire matrix. I cannot think of another maximum.
Image Analyst
am 23 Mai 2013
I haven't downloaded the data yet, but if it is a 2D array, then what are you plotting? Just one of the columns or one of the rows? Or you just plotted the whole thing and were unaware of what it would plot? What do you want to plot? A row or a column, or both, perhaps with surf() or waterfall()?
Ricky
am 23 Mai 2013
José-Luis
am 23 Mai 2013
Then you get the maximums with what Image Analyst suggested. Please look at the documentation for max().
If you want the row, columns of each column maximum, then you need to transform the linear indexes to subindexes. Please look at ind2sub().
If you want to plot a surface (2d data) then you can't use plot(). Please look at the documentation for surf(). There you will find links to similar plotting tools, that could also work.
Kategorien
Mehr zu Annotations finden Sie in Hilfe-Center und File Exchange
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!