how to convert signal into 2D image gray scale

hello ,how to convert signal into 2D image gray scale by CWT

Answer 1

[Ameer Hamza]

The documentation shows that if you call cwt(): https://www.mathworks.com/help/wavelet/ref/cwt.html with an output argument, then it will return a 2D matrix. You can take the absolute value of the matrix and display it as a grayscale image.

[ahmed Altameemi]

Thank you for reply , when I use it I got this error

Error using mat2gray

Expected input number 2, LIMITS, to be an array with number of elements equal to 2.

[Ameer Hamza]

The output of cwt should already by 2D matrix. You don't need to call mat2gray. You might need to rescale it to display it properly

M = cwt(input_signal);
M = abs(M);
M = rescale(M, 0, 1);
imshow(M)

[ahmed Altameemi]

but in this case I will get color image not gray

[Ameer Hamza]

I don't have the toolbox so cannot run the code, but can you show the output of

M = cwt(input_signal);
M = abs(M);
size(M)

If it is a 2D matrix, then imwhow will automatically show a grayscale image.

[Walter Roberson]

mat2gray() is an older routine that is suitable for rescaling data to the range [0 1]

However, you would not typically pass a second argument to mat2gray, and when you do pass a second argument it must be a vector of two values that is the range to scale to.

imshow() of a 2D array will not be color, but you might want to

imshow(M, [])

[ahmed Altameemi]

Thank you for reply Dear @Ameer Hamza I got this error

Error using cwt (line 300)

Expected input to be a vector.

load('ECGData.mat');
data=ECGData.Data;
ARR=data(1:30,:);
M = cwt(ARR);
M = abs(M);
size(M)

[ahmed Altameemi]

my main code

this,

%program to create CWT Image database from ECG signals 
load('ECGData.mat'); %load ECG data
data=ECGData.Data;    %Get Signal Values in data
labels =ECGData.Labels ;  %Get Labels in labels 
ARR=data(1:30,:); %Taken first 30 recordings 
CHF=data(97:126,:);
NSR=data(127:156,:);
signallength=500;
%Defining filters for CWT with amor wavelet and 12 filters per octave 
fb=cwtfilterbank('signalLength',signallength,'Wavelet','amor','VoicesPerOctave',12);
mkdir('ecgdataset'); %Main Folder 
mkdir('ecgdataset\arr'); %Sub Folder
mkdir('ecgdataset\chf');
mkdir('ecgdataset\nsr');
ecgtype={'ARR','CHF','NSR'};
%Function to convert ECG To Image

ecg2cwtscg(ARR,fb,ecgtype{1});
ecg2cwtscg(CHF,fb,ecgtype{2});
ecg2cwtscg(NSR,fb,ecgtype{3});

>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>

this is the fun

function ecg2cwtscg(ecgdata,cwtfb,ecgtype)
nos=10;%number of signals 
nol=500; %signal length 
colormap=jet(128);
if ecgtype=='ARR'
    folderpath=strcat('D:\ecgdataset\arr\');
    findx=0;
    for i=1:30
        indx=0;
        for k=1:nos
            ecgsignal=ecgdata(i,indx+1:indx+nol);
            cfs=abs(cwtfb.wt(ecgsignal));
            im=ind2rgb(im2uint8(rescale(cfs)),colormap);
            filenameindex=findx+k;
            filename=strcat(folderpath,sprintf('%darr.jpg',filenameindex));
            imwrite(imresize(im,[227 227]),filename);
            indx=indx+nol;
        end
        findx=findx+nos;
    end
elseif ecgtype=='CHF'
    folderpath=strcat('D:\ecgdataset\chf\');
    findx=0;
    for i=1:30
        indx=0;
        for k=1:nos
            ecgsignal=ecgdata(i,indx+1:indx+nol);
            cfs=abs(cwtfb.wt(ecgsignal));
            im=ind2rgb(im2uint8(rescale(cfs)),colormap);
            filenameindex=findx+k;
            filename=strcat(folderpath,sprintf('%dchr.jpg',filenameindex));
            imwrite(imresize(im,[227 227]),filename);
            indx=indx+nol;
        end
        findx=findx+nos;
    end
    elseif ecgtype=='NSR'
    folderpath=strcat('D:\ecgdataset\nsr\');
    findx=0;
    for i=1:30
        indx=0;
        for k=1:nos
            ecgsignal=ecgdata(i,indx+1:indx+nol);
            cfs=abs(cwtfb.wt(ecgsignal));
            im=ind2rgb(im2uint8(rescale(cfs)),colormap);
            filenameindex=findx+k;
            filename=strcat(folderpath,sprintf('%dnsr.jpg',filenameindex));
            imwrite(imresize(im,[227 227]),filename);
            indx=indx+nol;
        end
        findx=findx+nos;
    end
end 

>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>

data set

https://github.com/mathworks/physionet_ECG_data/\

I got from this code color image

I need it grayscale imgae

could you help me

[Walter Roberson]

cwt() only accepts vectors, not images. For grayscale images use cwtft2()

https://www.mathworks.com/help/wavelet/ref/cwtft2.html

[ahmed Altameemi]

thank you sir for reply could you please check this https://www.mathworks.com/matlabcentral/answers/630368-how-to-convert-signal-into-2d-image-gray-scale#comment_1097449

[Walter Roberson]

If the code is designed for color images, then take your grayscale images and repmat(GRAYSCALE,[1 1 3]) to get a color image and process it through the code.

However, the code you posted is for processing 1D EEG signals, not for images of any kind.

Mathworks does not provide any equivalent of cwtfilterbank for 2D.

[ahmed Altameemi]

thank you gays I sloved it , I just replace the

ind2rgb to ind2gray

[ahmed Altameemi]

Hello , when I run the above code I got 3 channel for each image , my quastion how can I get 1 channel for each image

[Ameer Hamza]

You can use rgb2gray() to convert 3 channels to 1.

[ahmed Altameemi]

how can I use it with my code above , thank you

[ahmed Altameemi]

I used but my data sitl 3 channel