MATLAB: How to pass images to a cnn for analysis

Question

Hi folks,

I'm not very familiar with matlab so apologies for the obvious question, but how can I pass an image to my cnn to be analysed?

My cnn's code is as follows:

AnisotropyDatasetPath = fullfile(matlabroot,'Training', 'Anisotropy');
IsotropyDatasetPath = fullfile(matlabroot,'Training', 'Isotropy');
FillerDatasetPath = fullfile(matlabroot,'Training', 'Filler');
TrainingDatasetPath = fullfile(matlabroot,'Training');
imds = imageDatastore(TrainingDatasetPath, 'IncludeSubfolders',true,...
    'LabelSource','foldernames');
labelCount = countEachLabel(imds)
numTrainFiles = 999;
[imdsTrain,imdsValidation] = splitEachLabel(imds,numTrainFiles,'randomize');
layers = [
    imageInputLayer([227 227 3])
    
    convolution2dLayer(3,8,'Padding','same')
    batchNormalizationLayer
    reluLayer
    
    maxPooling2dLayer(2,'Stride',2)
    
    convolution2dLayer(3,16,'Padding','same')
    batchNormalizationLayer
    reluLayer
    
    maxPooling2dLayer(2,'Stride',2)
    
    convolution2dLayer(3,32,'Padding','same')
    batchNormalizationLayer
    reluLayer
    
    fullyConnectedLayer(3)
    softmaxLayer
    classificationLayer];
options = trainingOptions('sgdm', ...
    'InitialLearnRate',0.01, ...
    'MaxEpochs',4, ...
    'Shuffle','every-epoch', ...
    'ValidationData',imdsValidation, ...
    'ValidationFrequency',30, ...
    'Verbose',false, ...
    'Plots','training-progress');
net = trainNetwork(imdsTrain,layers,options);
YPred = classify(net,imdsValidation);
YValidation = imdsValidation.Labels;
accuracy = sum(YPred == YValidation)/numel(YValidation)

thanks!

Answer 1

Hey Teshan,

After you train a neural network for deep learning using trainNetwork into net, follow this procedure:

• Use the function augmentedImageDatastore for effective preprocessing of the images
• Use the activations function to compute the DL Network layer activations i.e. the image features using the CNN net
• Use the predict function to finally make a prediction using the classifier

The overall workflow that I mentioned is available in this example as well.

Hope this helps!