getMask

Generate Signal and Visualize Spectrogram

Generate a signal that consists of a voltage-controlled oscillator and five Gaussian atoms. The signal is sampled at 14 kHz for two seconds. Plot the spectrogram of the signal.

Fs = 14000;
t = (0:1/Fs:2)';
st = 0.01;
gaussFun = @(A,x,mu,f) exp(-(x-mu).^2/(2*st^2)).*sin(2*pi*f.*x)*A';
atomTimeCenters = [0.2 0.5 1 1.3 1.8];
atomFreqCenters = [2 6 2 5 1]*1000;
atomAmplitudes = [1 1 1 1 1]/10;
s = gaussFun(atomAmplitudes,t,atomTimeCenters,atomFreqCenters) ...
    + vco(chirp(t+0.1,0,t(end),3).*exp(-2*(t-1).^2),[0.1 0.4]*Fs,Fs);

bt = 0.2;
tr = 0.05;
op = 99;
pspectrum(s,Fs,"spectrogram", ...
    Leakage=bt,TimeResolution=tr,OverlapPercent=op)

The spectrogram shows four patches in time-frequency domain that correspond with the Gaussian atoms. Define the times and frequencies for all the atoms.

atomTimes = atomTimeCenters'+[-st st]*5.5;
atomFreqs = atomFreqCenters'+[-1 1]*200;

Label Signal in Time-Frequency Domain

Create a numeric time-frequency ROI label definition to label the Gaussian atoms. Specify spectrogram options with leakage properties.

opts = labelSpectrogramOptions("leakage", ...
    Leakage=40*(1-bt),Overlap=op, ...
    TimeResolutionMode="specify",TimeResolution=tr);

lblDef = signalLabelDefinition("Atom", ...
    LabelDataType="numeric", ...
    LabelType="roiTimeFrequency",TimeFrequencyOptions=opts);

Create a labeled signal set from the signal and time-frequency ROI label definition.

lss = labeledSignalSet(s,lblDef,SampleRate=Fs);

Label the atoms in time-frequency domain. Set the label values to 1 through 5.

setLabelValue(lss,1,"Atom",atomTimes,atomFreqs,1:5)

Display the label table. Each label has its time-frequency ROI limits and a numeric value.

lblTable = lss.Labels.Atom{1};
lblTable.Value = cell2mat(lblTable.Value)

lblTable=5×3 table
    TimeROILimits     FrequencyROILimits    Value
    ______________    __________________    _____

    0.145    0.255       1800    2200         1  
    0.445    0.555       5800    6200         2  
    0.945    1.055       1800    2200         3  
    1.245    1.355       4800    5200         4  
    1.745    1.855        800    1200         5  

Get and Visualize Array of Label Masks

Create a time-frequency mask with the label table and spectrogram options. Specify the signal sample rate and mask format.

msk = timeFrequencyMask(lblTable,opts, ...
    SampleRate=Fs,MaskFormat="pixel",OverlapAction="prioritizeByList");

Read the time-frequency mask and get the array of label regions as a per-pixel matrix and as a 3-D binary array.

lblPerPixel = getMask(msk,s,PixelType="perpixel");
lblBinaries = getMask(msk,s,PixelType="binary");

Plot the array of labels in per-pixel format. The white rectangular patches represent time-frequency regions where a label exists.

figure
tiledlayout("flow")
nexttile
imshow(~isnan(lblPerPixel))
title("2-D per-pixel mask (grayscale image)")

Plot the first three pages of the array of labels in binary format. Each page consists of pixels marked with true (1) or false (0), where true (1) represents a labeled time-frequency ROI. The three-page 3-D array forms an image with colored rectangular patches that depict the labeled ROIs:

nexttile
imshow(double(lblBinaries(:,:,1:3)))
title("3-D binary mask (RGB image)")