How to generate Image from Raw Data and find Peak Value

Hello, I have the following Dataset in which first column shows the X-axis and 2nd Column shows the Y-axis.

I want to create Image from the Dataset and find Peak Value as shown in the image below and

Save the X and Y axis Values Corresponding Peak.

How can i do that in MATLAB

Answer 1

[Nikhilesh]

For your case it would be better if you can pronounce i want to creat a plot/graph out of a dataset instead of image. For plotting the data you can use the following code.

plot(Dataset(:,1),Dataset(:,2));

Plese have a look at the "max" function. Which you can use to find the peak

https://www.mathworks.com/help/matlab/ref/max.html?searchHighlight=max&s_tid=srchtitle_max_1

peak_value = max(Dataset(:,2)); 
peak_index = find(Dataset(:,2) == peak_value);
peak_x = Dataset(peak_index, 1);
peak_y = Dataset(peak_index, 2);

[Med Future]

@Nikhilesh

In the Following Dataset I have two peaks. Can you please modified the code for this.

For example. find the first peak and its X and Y values. then Find the 2nd peak and its X and Y Values.

And Subtract the 2nd X value with 1st X value.

Answer 2

[Mathieu NOE]

hello

try this code below

the peak coordinates are stored in xmax and ymax

for one single peak , max is sufficient; if you need to identify multiple peaks you can use findpeaks , islocalmax, islocalmin

load('DatasetCir.mat')

x = Dataset(:,1);
y = Dataset(:,2);

% find max value 
[ymax,idx] = max(y);
xmax = x(idx);
plot(x,y,'o',xmax,ymax,'*r');
text(xmax*1.005,ymax,['x = ' num2str(xmax)]);
text(xmax*1.005,ymax*0.95,['y = ' num2str(ymax)]);

[Med Future]

@Mathieu NOE In the Following Dataset I have two peaks. Can you please modified the code for this.

For example. find the first peak and its X and Y values. then Find the 2nd peak and its X and Y Values.

And Subtract the 2nd X value with 1st X value.

[Mathieu NOE]

hello again

here a solution based on findpeaks (many arguments you can use to tweak the code to your data)

load('twopeaksdata.mat')
x = Twopeaksdata(:,1);
y = Twopeaksdata(:,3);
% find max value 
[ymax,idx] = findpeaks(y,'NPeaks',2,'SortStr','descend');
xmax = x(idx);
% Subtract the 2nd X value with 1st X value.
deltaX = xmax(2) - xmax(1)

plot(x,y,'o',xmax,ymax,'*r');
for ck = 1:numel(xmax)
text(xmax(ck)+0.05*max(x),ymax(ck),['x = ' num2str(xmax(ck))]);
text(xmax(ck)+0.05*max(x),ymax(ck)*0.95,['y = ' num2str(ymax(ck))]);
end

[Med Future]

@Mathieu NOE Thanks for the answer. But you have given an argument'NPeaks',2.

But Sometime I have not known How many Peaks How can i solve that problem?

[Mathieu NOE]

Well findpeaks cannot tell it by itself how many peaks you want, it's powerfull but not magical

here you told me you wanted the two dominant peaks and that's how I coded it

simply plot your data first and decide how you want to use findpeaks

look at the doc and examples and see how to use the different options (you can combine them)

[Med Future]

@Mathieu NOE Thanks for your reply. But i want an algo which Finds the peaks in the Dataset Like how many peaks there.

and find the maximum Value of Y axis and its respected X axis Value.

[Mathieu NOE]

the definition of "peaks" can vary from one data to another and what you are looking for

for example one person is interested to find not only the dominant peaks but also some local maxima

like on this plot

so this person want all green points , but you don't want this maybe

that's why there is no single method that covers all needs ; somehow you need to tell the algo what you want

this is why findpeaks does have many options , you need to learn how to use it

I can show you an alternative , but here again you must give two parameters to get the right output

again, there is no magical code that can detect in your brain what you want and how to proceed

you can look at the File Exchange and pick one function that may better serve your needs (here I choose : PeakSeek - File Exchange - MATLAB Central (mathworks.com) )

load('twopeaksdata.mat')
x = Twopeaksdata(:,1);
y = Twopeaksdata(:,3);

% find max value 
% [ymax,idx] = findpeaks(y,'NPeaks',2,'SortStr','descend');
% [ymax,idx] = findpeaks(y,'MinPeakHeight',max(y)/2,'SortStr','descend');

 % alternative with peakseek (Fex : 
minpeakdist = 1000; % 
minpeakh = max(y)/10;
[idx, ymax]=peakseek(y,1000,max(y)/2);
xmax = x(idx);
% Subtract the 2nd X value with 1st X value.
deltaX = xmax(2) - xmax(1)

plot(x,y,'o',xmax,ymax,'*r');
for ck = 1:numel(xmax)
text(xmax(ck)+0.05*max(x),ymax(ck),['x = ' num2str(xmax(ck))]);
text(xmax(ck)+0.05*max(x),ymax(ck)*0.95,['y = ' num2str(ymax(ck))]);
end

% Fex submission : 
% https://fr.mathworks.com/matlabcentral/fileexchange/26581-peakseek?s_tid=srchtitle

function [locs, pks]=peakseek(x,minpeakdist,minpeakh)
% x is a vector input (generally a timecourse)
% minpeakdist is the minimum desired distance between peaks (optional, defaults to 1)
% minpeakh is the minimum height of a peak (optional)
%
% (c) 2010
% Peter O'Connor
% peter<dot>ed<dot>oconnor .AT. gmail<dot>com
if size(x,2)==1, x=x'; end
% Find all maxima and ties
locs=find(x(2:end-1)>=x(1:end-2) & x(2:end-1)>=x(3:end))+1;
if nargin<2, minpeakdist=1; end % If no minpeakdist specified, default to 1.
if nargin>2 % If there's a minpeakheight
    locs(x(locs)<=minpeakh)=[];
end
if minpeakdist>1
    while 1
        del=diff(locs)<minpeakdist;
        if ~any(del), break; end
        pks=x(locs);
        [garb, mins]=min([pks(del) ; pks([false del])]); %#ok<ASGLU>
        deln=find(del);
        deln=[deln(mins==1) deln(mins==2)+1];
        locs(deln)=[];
    end
end
if nargout>1
    pks=x(locs);
end
end

[Mathieu NOE]

here another altrenative with

this is very simple to use as this function only ask you the number of peaks to find out

it differs from more complex functions like findpeaks (with the many options)

load('twopeaksdata.mat')
x = Twopeaksdata(:,1);
y = Twopeaksdata(:,3);

% find max value 
% [ymax,idx] = findpeaks(y,'NPeaks',2,'SortStr','descend');
% [ymax,idx] = findpeaks(y,'MinPeakHeight',max(y)/2,'SortStr','descend');

 % alternative with peakfind 
 Np = 2;
[PeakData]=peakfind(x,y,Np); % PeakData is 3 columns array : data indice , x , y values
xmax = PeakData(:,2);
ymax = PeakData(:,3);

% Subtract the 2nd X value with 1st X value.
deltaX = xmax(2) - xmax(1)

plot(x,y,'o',xmax,ymax,'*r');
for ck = 1:numel(xmax)
text(xmax(ck)+0.05*max(x),ymax(ck),['x = ' num2str(xmax(ck))]);
text(xmax(ck)+0.05*max(x),ymax(ck)*0.95,['y = ' num2str(ymax(ck))]);
end


  function [PeakData]=peakfind(Xdata,Ydata,NumPeaks)
% function [PeakData]=peakfind(Xdata,Ydata,NumPeaks)
% Returns array PeakData with NumPeaks entries consisting of 
% the xindex,xvalue,yvalue
% If the Ydata is a complex array, look at the magnitude.
% 

% make sure data is row oriented (MN)
Xdata = (Xdata(:))';
Ydata = (Ydata(:))';


   iYpeak    = 3;
   PeakState = [1 1 -inf];
   PeakData  = [];
   ibeg = 1;    % ignore DC line
   iend = length(Ydata);
   if ~isreal(Ydata)
         Ydata = abs(Ydata);
   end;
   
   for i = 1:NumPeaks
      ix=find(Ydata(ibeg:iend) > [-inf, Ydata(ibeg:iend-1)] & ...   % must be > point on left
              Ydata(ibeg:iend) > [Ydata(ibeg+1:iend), -inf] & ...   % must be > point on right
              Ydata(ibeg:iend) < PeakState(iYpeak));                % must be < previous peak
      if isempty(ix) 
         [y , ix] = max(Ydata);   % no more peaks: go to max
      else
         [y , i]  =  max(Ydata(ix));  % there are more: go to biggest
         ix     = ix(i);   
      end;
      
      PeakState= [ix,Xdata(ix),y];
      PeakData = [PeakData;PeakState];    
   end